Preface Motivation In today's information age, computers are being used in every walk of life. They are being used by people of all age and profession in their work and in their leisure. This new social revolution has changed the basic concept of 'Computing'. Computing in today's information age is no more limited to computer programmers and computer engineers. It has become an activity of a common man. Rather than knowing how to program a computer, most computer users simply need to understand how a computer functions and what all it can do. Even those who need to program a computer can do their job more effectively with a better understanding of how computers function and the capabilities and limitations of computers. As a result, almost all academic institutions have started offering regular courses on foundations of computing at all levels. These courses deal with the fundamental concepts of the organization, functions, and usage of modern computer systems. Hence we realized that a good textbook that can cover these concepts in an orderly manner would certainly be very useful for a very wide category of students and all types of computer users. Motivated by these factors, we decided to do research towards the preparation of a textbook on foundations of computing. Our primary objective was to concisely present a clear explanation of the current state of the art in computing so that after reading and understanding the material in this text reader will be ready to play an active role in the computer age. Audience This book is intended for anyone who is interested in knowing about computers. It is designed to meet the needs of beginners as well as advanced learners who wish to deal with computers. More specifically, different types of audience can benefit from this book in the following manner: • As knowledge of computers has become important for people from all disciplines, this book can be used by students of computer science, all disciplines of engineering, business administration, management, general science, commerce and arts as well. • At the undergraduate level, where an introductory course on computers is included as a part of the curriculum, this book can be used as a textbook. • The book can also serve as a good course material for participants of short-term courses conducted by various training organizations. • As the book explains the computer jargons in a simple language, it can also be used by various types of computer users as a self-study material for obtaining a basic understanding of the various hardware and software components of computers. • The coverage of many advanced topics and current state of the art in computing makes the book very useful and informative for programmers and other computer professionals as well.
In short, this book is for everyone who is either excited about computers or interested in knowing about computers. Knowledge of computers is not prerequisite to follow the contents of the book. It assumes no background in computers or computer programming. Overview The layout and organization of the book and its content are designed to present the fundamental concepts of computers in an interesting, logical, and informative manner. The book does not concentrate on any particular computer system or hardware. Instead, it discusses, in a general setting, the organization, functions, and usage principles that are applicable to all types of modern computer systems. Starting with an introductory chapter, the book goes on to cover all facets of computers. It uses the simple-to-complex and easy-to-learn approach throughout. Special attention has been paid to remove unwanted details and to make the material comprehensive, concise and practical. Simple and plain English has been used throughout. All the concepts presented in the book are illustrated with suitable examples as and when required. The book also contains numerous diagrams for better illustration. These features will enable the readers to grasp the presented concepts easily and to use the book effectively for self-study. Each chapter contains a section on 'Points to Remember' that highlights the important points covered in the chapter. Each chapter also contains a large number of carefully designed review questions that are meant to test the understanding of the materials presented in the chapter. To make the text easy to read and understand, the style of presentation used throughout the book is motivational, non-mathematical and explanatory in nature. Contents The book begins with an introduction to computers. Chapter 1 explains what this versatile tool is, why it is so powerful and useful, what its limitations are, and how it has evolved from its early days to become a powerful and useful tool for all types of users in today's society. Chapter 2 presents the basic organization of a computer system. It briefly explains about the five main units present in any computer system that correspond to the five basic operations performed by all computer systems. Chapter 3 introduces the basic fundamentals of number system and some of the number systems commonly used by computer professionals. Chapter 4 describes how binary
numbers are coded to represent characters in computers and some of the most commonly used computer codes. Chapter 5 describes how the basic arithmetic operations are performed inside the computer using binary numbers. Chapter 6 introduces Boolean algebra, which is algebra to deal with the binary number system. This chapter also describes logic gates, which are used for building logic circuits used by the processors of computer systems. Chapter 7, 8 and 9 describe the internal structure and working of the various components of a computer system. Chapter 7 deals with the structure and working of the Central Processing Unit (popularly known as CPU or processor) and the primary storage (popularly known as main memory or simply memory). Chapter 8 deals with the structure and working of the various types of secondary storage devices, and Chapter 9 deals with the structure and working of the various types of input/output devices. Chapter 10 provides an introduction to the software of computers. It explains why both hardware and software are necessary for a computer to do useful job, what are the various types of software, and how to acquire or develop software. Chapters 11, 12, and 13 describe the various types of commonly used tools and techniques for planning, development, implementation, and operation of software systems. Chapter 11 deals with the commonly used tools for planning computer programs before starting to write them. These tools help in producing correct and effective computer programs. Chapter 12 deals with the commonly used programming languages for writing computer programs. It differentiates between high-level and low-level programming languages and briefly describes a number of popular high-level languages. It also explains how to select a programming language out of the many options available for coding an application. Chapter13 deals with the main activities of the implementation and operation phase, which include testing and debugging of programs, complete documentation of the system, changeover from an existing system to the new system, and system modification and maintenance. Chapter 14 presents the important concepts of operating systems, which is a system software present in all computer systems. It also briefly describes a number of popular operating systems. Chapter 15 describes some of the most commonly used application software packages. They are word-processing package, spreadsheet package, graphics package, and personal assistance package. Chapter 16 deals with business data processing. It describes the basic concepts of database systems, and the modern concepts in database technology. Chapter 17 presents the important concepts of data communications, computer networks, and distributed computing systems.
Chapter 18 presents a description of the Internet. It explain what it is, how it has evolved, what are the basic services offered by it, what are the various types of tools available to make effective use of the Internet, what are the common uses of the Internet and how to get connected to the Internet. Chapter 19 deals with multimedia computing. It describes the different types of data (text, graphics, animation, audio, and video) that modern computer systems can process and the issues involved in processing these data types. Chapter 20 explains how computers are classified as notebook computers, personal computers, workstations, mainframe systems, and supercomputers. It also describes client-server computing and explains what is a client computer and a server computer. Acknowledgement A book of this type naturally gained a number of ideas from previous books dealing with the subject matter presented in this book. The authors express their thanks to all these authors, too numerous to acknowledge individually. We are thankful to IEEE PRESS for granting us permission to use in this book some material from the book entitled 'Distributed Operating Systems: Concepts and Design' written by the first author and published by IEEE PRESS. We are thankful to Mr. R. K. Arora, Executive Director, C-DAC for granting us permission to use the photograph of PARAM Supercomputing system in the book. We are also thankful to Mr. R. K. Arora and Mr. S. P. Dixit of C-DAC for their continuous encouragement and support throughout the course of this project. We would like to thank Ms. Sushma Pawar for her support activities. Her hard work and cheerful devotion to the project has helped us in bringing out the book in a timely manner. We would also like to thank Mr. Rajeev Rudrakshi for his excellent design of the cover page. We are also thankful to Mr. Manish Jain and other members of BPB Publications for their support and help in publishing the work. Finally, we would like to thank our eight years old son, Deeptansu, for his loving support and understanding during the entire period of this long project. P. K. Sinha Priti Sinha August, 2002 List of Abbreviations
4GLs
Fourth-Generation Languages
C-DAC
Centre for Advanced
Development
of
A/D ABC AI ALGOL ALU AM ANSI AOL ASCII
Analog-to-Digital Computing Atanasoff-Berry Computer CDC Control Data Corporation Artificial Intelligence CDMA Code-Division-Multiple-Access ALGOorithmic Language CDPD Cellular Digital Packet Data Arithmetic Logic Unit CD-ROM Compact Disk-Read Only Memory Amplitude Modulation CD-R CD-Recordable American National Standards Institute CGA Color Graphics Adapter America OnLine CISC Complex Instruction Set Computer American Standard Code for CL Command Language Information Interchange CLV Constant Linear Velocity ATM Automatic Teller Machine; COBOL COmmon Business Asynchronous Transfer Mode Oriented Language BARC Bhabha "Atomic Research Centre CODASYL Conference On DAta SYstems BASIC Beginners All-purpose Symbolic Languages Instruction Code COM Computer Output Microfilm; BCD Binary Coded Decimal Component Object Model BIOS Basic Input Output System CPS Characters Per Second BIPS Billions of Instructions Per Second CPU Central Processing Unit B-ISDN Broad-band ISDN CRT Cathode-Ray Tube BLOBs Binary Large OBjects Cscw Computer Supported Cooperative BNA Burroughs Network Architecture Working BOT Beginning-Of-Tape D/A Digital-to-Analog BPI Bytes Per Inch DAB Device Address Bus BSD Berkeley Software Distributions DAT Digital Audio Tape BSNL Bharat Sanchar Nigam Limited DBMS DataBase Management System CAD Computer-Aided Design DCBR Dynamic Content Based Retrieval CAE Computer-Aided Engineering DCT Discrete Cosine Transform CAM Computer-Aided Manufacturing DDL Data Definition Language CASE Computer-Aided Software DDS Digital Data Storage Engineering DEC Digital Equipment Corporation CAV Constant Angular Velocity DES Data Encryption Standard CBR/CBQ Content-Based Retrieval/Querying DMA Direct Memory Access CCITT Consultative Committee for DML Data Manipulation Language International Telephony and DNA Digital Network Architecture Telegraphy DoD U.S. Department of Defense DRDO Defense Research and Development HD-MAC High Definition Multiple Analog Organization Components DSDD Double-Sided Double Density HDTV High Definition Television DSHD Double-Sided High Density HP Hewlett Packard DSN Distributed Systems Network HSCSD High-Speed-Circuit-Switched-Data DVD Digital Versatile Disk HTML HypterText Markup Language EBCDIC Extended Binary-Coded Decimal HTTP HyperText Transport Protocol Interchange Code -• Hz Hertz ECA-rule Event-Condition-Action-rule I/O Input and Output EDGE Enhanced Data-rate for Global IBG Inter-Block Gap Evolution IBM International Business Machines
EDSAC EDVAC
Electronic Delay Storage Automatic Calculator Electronic Discrete Variable
Automatic Computer EEPROM Electrically EPROM EGA Extended Graphics Adapter EISA Bus Extended Industry Standard Architecture Bus ELD Electro-Luminescent Display ENIAC Electronic Numerical Integrator And Calculator EOT End-Of-Tape EPIC Explicitly Parallel Instruction Computing EPROM Erasable Programmable Read-Only Memory ERNET Education and Research NETwork ESDI Enhanced Small Device Interface ESMR Extended Specialized Mobile Radio ETSI European Telecommunications FAT FDM FEP FLOPS FM FMS FORTRA N FSK FTP GB GEO GFLOPS GIGO GIS GPD GPRS GSM
IC IDE iDEN IDN IEEE IETF IMT-2000 IP IPI-3 IPV4 IPV6 IRG ISA ISA Bus ISAM ISDN ISO
Standard Institute File Allocation Table Frequency-Division Multiplexing Front-End Processor Floating Point Operations Per Second Frequency Modulation File Management System FORmula TRANslation
ISP ITU
Frequency Shift Keying File Transfer Protocol Giga Bytes Geosynchronous-Earth-Orbit Giga FLOPS
LCD LED LEO LISP LMDS
Garbage-In-Garbage-Out Geographical Information System Gas-Plasma Display General Packet Radio Service Global System for Mobile communication
LPM LQ LRU LSB LSD T
JCL JPEG KB KDD LAN
Integrated Circuit Integrated Drive Electronics Integrated Digital Enhanced Network Integrated Digital Networks Institute of Electrical and Electronics Engineers Internet Engineering Task Force International Mobile Telecommunications 2000 Internet Protocol Intelligent Peripheral InterfaceLevel 3 Internet Protocol Version 4 Internet Protocol Version 6 Inter-Record Gap Instruction Set Architecture Industry Standard Architecture Bus Indexed Sequential Access Method Integrated Services Digital Network International Standards Organization Internet Service Provider (ISP) International Telecommunication Union Job Control Language Joint Photographic Experts Group Kilo Bytes Knowledge Discovery in Databases Local Area Network Liquid Crystal Display Light-Emitting Diode Low-Earth-Orbit LISt Processing Local Multipoint Distribution System Lines Per Minute Letter Quality Least Recently Used Linux Standard Base Least Significant Digit
TCP
Transport Control Protocol
VTOC
Volume Table Of Contents
TCP/IP
Transmission Control Protocol/Internet Protocol Time-Division CDMA
WAE WAN WAP
Wireless Application Environment Wide Area Network Wireless Application Protocol
Time-Division Multiplexing Time-Division-Multiple-Access Typeset Quality * *• User Datagram Protocol Ultra Large Scale Integration UNIVersal Automatic Computer Universal Product Code Uniform Resource Locator Unshielded Twisted-Pair Ultra Violet EPROM
W-CDMA
Wideband Code-Division Multiple Access Wireless Datagram Protocol Wireless Local Area Network Wireless Local Loop Wireless Markup Language Write-Once Read Many Wireless Session Protocol What You See Is What You Get Wireless Transport Layer Security
Value Added Network Video Cassette Recorder Video Display Terminal Video Graphics Array Very Large Scale Integration Video-On-Demand Voice Over IP Very Small Aperture Terminals Videsh Sanchar Nigam Limited
WTP WWW X.400 X.500 XDR XHTML
TDCDMA TDM TDMA TQ UDP ULSI UNIVAC UPC URL UTP UVEPRO M VAN VCR VDT VGA VLSI VOD VOIP VSAT VSNL
WDP WLAN WLL WML WORM WSP WYSIWYG WTLS
XML
Wireless Transaction Protocol Word Wide Web Electronic Mail Protocol Directory Server Protocol external Data Representation extensible HyperText Markup Language extensible Markup Language
Chapter 1 Introduction The word "computer" comes from the word "compute" which means to calculate. So a computer is normally considered to be a calculating device that can perform arithmetic operations at enormous speed. In fact, the original objective for inventing the computer was to create a fast calculating machine. But more than 80% of the work done by computers today is of nonmathematical or non-numerical nature. Hence, to define a computer merely as calculating device is to ignore over 80% of its work. More accurately, a computer may be defined as a device that operates upon data. Data can be anything like bio-data of various applicants when the computer is used for recruiting personnel, or the marks obtained by various students in various subjects when the computer is used to prepare results, or the details (name, age, sex, etc.) of various
passengers when the computer is employed for making airline or railway reservations, or numbers of different types in case of application of computers for scientific research problems, etc. Thus, data comes in various shapes and sizes depending upon the type of computer application. A computer can store, process, and retrieve data as and when desired. The fact that computers process data is so fundamental that many people have started calling it a data processor. The name data processor is more inclusive because modern computers not only compute in the usual sense but also perform other functions with the data that flow to and from them. For example, data processors may gather data from various incoming sources, merge (process of mixing or putting together) them all, sort (process of arranging in some sequence - ascending or descending) them in the desired order and finally print them in the desired format. None of these operations involve the arithmetic operations normally associated with the computing device but the term computer is often applied anyway. The activity of processing data using a computer is called data processing. Data processing consists of three sub-activities: capturing the input data, manipulating the data, and managing the output results. As used in data processing, information is data arranged in an order and form that is useful to the people who receive it. That is, information is relevant knowledge, produced as the output of data processing operation and acquired by people to enhance understanding and to achieve specific purposes. Thus data is the raw material used as input to data processing and information is the processed data obtained as the output of data processing. CHARACTERISTICS OF COMPUTERS The increasing popularity of computers has proved that it is a very powerful and useful tool. The power and usefulness of this popular tool ate mainly due to its following characteristics: 1. Automatic. A machine is said to be automatic if it works by itself without human intervention. Computers are automatic machines because once started on a job, they carry on until the job is finished, normally without any human assistance. However, computers being machines, cannot start themselves. They cannot go out and find their own problems and solutions. They have to be instructed. That is, a computer works from a program of coded instructions that specify exactly how a particular job is to be done. While the job is in progress, the program is stored in the computer, and the parts of the instructions are obeyed. As soon as one instruction is completed, the next is obeyed automatically. Some of the other characteristics of computers such as speed and accuracy are due to the fact that they are automatic and work on a problem without any human intervention. 2. Speed. A computer is a very fast device. It can perform in a few seconds the amount of work that a human being can do in an entire year - if he worked day and night and did
nothing else. To put it in a different manner, a computer does in one minute what would take a man his entire lifetime. While talking about the speed of a computer, we do not talk in terms of seconds or even milliseconds (10'3). Our units of speed are the microseconds (10'6), the nanoseconds (10'9), and even the . picoseconds (10'12). A powerful computer is capable of performing several billion (109) simple arithmetic operations per second. 3. Accuracy. In addition to being very fast, computers are very accurate. The accuracy of a computer is consistently high and the degree of accuracy of a particular computer depends upon its design. But for a particular computer, each and every calculation is performed with the same accuracy. The electronic circuits of a computer that perform computations have no mechanical parts to wear and malfunction, resulting in errorless operation continuously. Hence, if the input data entering the computer are correct and if the program of instructions is reliable, then we can expect that the computer generally will produce accurate output. Errors can occur in a computer, but these are mainly due to human rather than technological weaknesses, that is, due to imprecise thinking by the programmer (a person who writes instructions for a computer to solve a particular problem) or due to incorrect input data. Computer errors caused due to incorrect input data or unreliable programs is often referred to as garbagein-garbage-out or GIGO. 4. Diligence. Unlike human beings, a computer is free from monotony, tiredness, lack of concentration, etc., and hence can work for hours together without creating any error and without grumbling. Due to this property, computers obviously score over human beings in doing routine type of jobs, which require great accuracy. If ten million calculations have to be performed, a computer will perform the ten millionth calculation with exactly the same accuracy and speed as the first one. 5. Versatility. Versatility is one of the most wonderful things about the computer. One moment, it is preparing the results of particular examination, the next moment it is busy preparing electricity bills, and in between, it may be helping an office secretary to trace an important letter in seconds. All that is required to change its talent is to slip in a new program (a sequence of instructions for the computer) into it. Briefly, a computer is capable of performing almost any task provided that the task can be reduced to a series of logical steps. 6. Power of Remembering. As a human being acquires new knowledge, the brain subconsciously selects what it feels to be important and worth retaining in its memory, and relegates unimportant details to the back of the mind or just forgets them. With computers, that is not the case. A computer can store and recall any amount of information because of its secondary storage (a type of detachable memory) capability. Every piece of information can be retained as long as desired by the user and can be recalled as and when required. Even after several years, the information recalled would be as accurate as on the day when it was fed to the computer. A computer forgets or
looses certain information only when it is asked to do so. So it is entirely up to the user to make a computer retain or forget a particular information. 7. No I. Q. A computer is not a magical device. It can only perform tasks that a human being can. The difference is that it performs these tasks with unthinkable speed and accuracy. It possesses no intelligence of its own. Its I. Q. is zero, at least till today. It has to be told what to do and in what sequence. Hence, only the user can determine what tasks a computer will perform. A computer cannot take its own decision in this regard. 8. No Feelings. Computers are devoid of emotions. They have no feelings and no instincts because they are machines. Although men have succeeded in building a memory for the computer, but no computer possesses the equivalent of a human heart and soul. Based on our feelings, taste, knowledge, and experience, we often make certain judgements in our day-to-day life. But computers cannot make such judgements on their own. Their judgement is based on the instructions-given to them in the form of programs that are written by us. They are only as good as man makes and uses them. THE EVOLUTION OF COMPUTERS Necessity is the mother of invention. The saying holds true for computers also because computers were invented as a result of man's search for fast and accurate calculating devices. The earliest device that qualifies as a digital computer is the "abacus" also known as "soroban". This device (shown in Figure 1.1) permits the users to represent numbers by the position of beads on a rack. Simple addition and subtraction can be carried out rapidly and efficiently by positioning the beads appropriately. Although, the abacus was invented around 600 B.C., it is interesting to note that it is still used in the Far East and its users can calculate at amazing speeds. Another manual calculating device was John Napier's bone or cardboard multiplication calculator, designed in the early 17th century and its upgraded versions were in use even around 1890. The first mechanical adding machine was invented by Blaise Pascal in 1642. Later, in the year 1671, Baron Gottfried Wilhelm von Leibniz of Germany invented the first calculator for multiplication. Keyboard machines originated in the United States around 1880 and are extensively used even today. Around this period only, Herman Hollerith came up with the concept of punched cards, which were extensively used as input media in computers even in late 1970s. Business machines and calculators made their appearance in Europe and America towards the end of the nineteenth century. Charles Babbage, a nineteenth century Professor at Cambridge University, is considered to be the father of modern digital computers. During his period, mathematical and statistical tables were prepared by a group of clerks. Even the utmost care and
precautions could not eliminate human errors. Babbage had to spend several hours checking these tables. Soon he became dissatisfied and exasperated with this type of monotonous job. The result was that he started thinking to build a machine, which could compute tables guaranteed to be error-free. In this process, Babbage designed a "Difference Engine" in the year 1822 which could produce reliable tables. In 1842, Babbage came out with his new idea of Analytical Engine that was intended to be completely automatic. It was to be capable of performing the basic arithmetic functions for any mathematical problem and it was to do so at an average speed of 60 additions per minute. Unfortunately, he was unable to produce a working model of this machine mainly because the precision engineering required to manufacture the machine was not available during that period. However, his efforts established a number of principles, which have been shown to be fundamental to the design of any digital computer. In order to have a better idea of the evolution of computers, let us now briefly discuss about some of the well-known early computers. These are as follows: 1. The Mark I Computer (1937-44). Also known as Automatic Sequence Controlled calculator, this was the first fully automatic calculating machine designed by Howard A. Aiken of Harvard University in collaboration with IBM (International Business Machines) Corporation. Its design was based on the techniques already developed for punched card machinery. Although this machine proved to be extremely reliable, it was very complex in design and huge in size. It used over 3000 electrically actuated switches to control its operations and was approximately 50 feet long and 8 feet high. It was capable of performing five basic arithmetic operations: addition, subtraction, multiplication, division and table reference. A number as big as 23 decimal digits could be used in this machine. It took approximately 0.3 second to add two numbers and 4.5 seconds for multiplication of two numbers. Hence, the machine was very slow as compared to today's computers. It was basically an electro-mechanical device since both mechanical and electronic components were used in its design. Although its operations were not controlled electronically, Aiken's machine is often classified as computer because its instructions, which were entered by means pf punched paper tape, could be altered. 2. The Atanasoff-Berry Computer (1939-42). This electronic machine w is developed by Dr. John Atanasoff to solve certain mathematical equations. It was called the Atanasoff-Berry Computer, or ABC, after its inventor's name and his assistant, Clifford Berry. It used 45 vacuum tubes for internal logic and capacitors for storage. 3. The ENIAC (1943-46). The Electronic Numerical Integrator And Calculator (ENIAC) was the first' all electronic computer. It was constructed at the Moore School of Engineering of the University of Pennsylvania, U.S.A. by a design team led by Professors J. Presper Eckert and John Mauchly.
ENIAC was developed as a result of military need. It took up the wall space in a 20 x 40 square feet room and used 18,000 vacuum tubes. The addition of two numbers was achieved in 200 microseconds, and multiplication in 2000 microseconds. Although, much faster in speed as compared to Mark I computer, ENIAC had two major shortcomings: it could store and manipulate only a very limited amount of information, and its programs were wired on boards. These limitations made it difficult to detect errors and to change the programs. Hence, its use was limited. However, whatever be the shortcomings of ENIAC, it represented an impressive feat of electronic engineering and was used for many years to solve ballistic problems. 4. The EDVAC (1946-52). The operation of ENIAC was seriously handicapped by the wiring board. This problem was later overcome by the new concept of "stored program" developed by Dr. John Von Neumann. The basic idea behind the stored program concept is that a sequence of instructions as well as data can be stored in the memory of the computer for the purpose of automatically directing the flow of operations. The stored program feature considerably influenced the development of modern digital computers and because of this feature we often refer to modern digital computers as stored program digital computers. The Electronic Discrete Variable Automatic Computer (EDVAC) was designed on stored program concept. Von Neumann has also got a share of the credit for introducing the idea of storing both instructions and data in the binary form (a system that uses only two digits - 0 & 1 to represent all characters) instead of the decimal numbers or human readable words. 5. The EDSAC (1947-49). Almost simultaneously with EDVAC of U.S.A., the Britishers developed the Electronic Delay Storage Automatic Calculator (EDSAC). The machine executed its first program in May 1949. In this machine, addition operation was accomplished in 1500 microseconds, and multiplication operation in 4000 microseconds. The machine was developed by a group of scientists headed by Professor Maurice Wilkes at the Cambridge University Mathematical Laboratory. 6. Manchester Mark I (1948). This computer was a small experimental machine based on the stored program concept. It was designed at Manchester University by a group of scientists headed by Professor M. H. A. Newman. Its storage capacity was only 32 words, each of 31 binary digits. This was too limited to store data and instructions. Hence, the Manchester Mark I was hardly of any practical use. 7. The UNIVAC I (1951). The Universal Automatic Computer (UNIVAC) was the first digital computer, which was not "one of a kind". Many UNIVAC machines were produced, the first of which was installed in the Census Bureau in 1951 and was used continuously for 10 years. The first business use of a computer, a UNIVAC I, was by General Electric Corporation in 1954. In 1952, the International Business Machines (IBM) Corporation introduced the 701 commercial computer. In rapid succession, improved models of the UNIVAC I and other
700-series machines were introduced. In 1953, IBM produced the IBM-650 and sold over 1000 of these computers. The commercially available digital computers, that could be used for business and scientific applications, had arrived. THE COMPUTER GENERATIONS "Generation" in computer talk is a step in technology. It provides a framework for the growth of the computer industry. Originally, the term "generation" was used to distinguish between varying hardware technologies. But nowadays, it has been extended to include both the hardware and the software, which together make up an entire computer system. The custom of referring to the computer era in terms of generations came into wide use only after 1964. There are totally five computer generations known till today. Each generation has been discussed below in detail along with their, identifying characteristics. Although there is a certain amount of overlapping between the generations, the approximate dates shown against each are normally accepted. During the description of the various computer generations, you will come across several terminologies and computer jargons, which you might not be aware of and may not be able to understand properly. Please do not worry about it because the idea here is to just give you an overview of the major developments and technologies during the five generations of computers. These developments and technologies will be described in greater detail in subsequent chapters. And remember that the objective of this book is also the same - to introduce you to the various concepts about computers. Hence you will have a better understanding of the terminologies introduced in this section only after you have completed reading this entire book. The objective of this section is mainly to provide an overview of what all you are going to learn in this entire book. First Generation (1942-1955) We have already discussed about some of the early computers - ENIAC, EDVAC, EDSAC, UNIVAC I, and IBM 701. These machines and others of their time were built by using thousands of vacuum tubes. A vacuum tube [see Figure 1.2(a)] was a fragile glass device that used filaments as a source of electrons and could control and amplify electronic signals. It was the only high-speed electronic switching device available in those days. These vacuum tube computers could perform computations in milliseconds and were referred to as first-generation computers. We also saw that the concept of stored program and the idea of storing both instructions and data in the binary form were introduced in 1946. Hence most of the first-generation computers worked on the principle of storing program instructions along with the data in the memory of the computer so that they could automatically execute a program without human intervention. The memory of these computers was constructed using
electromagnetic relays and all data and instructions were fed into the system from punched cards. Of course, the instructions were written in machine and assembly languages because high-level programming languages were introduced much later (computer languages are covered in Chapter 12). Even assembly language was invented in early fifties, so the first-generation computers of 1940s were programmed in machine language only. Only the first-generation computers introduced later such as UNIVAC I and IBM 701 could be programmed in assembly language. Because machine and assembly languages are very difficult to work with, only a few specialists understood how to program these early computers. Moreover, the concept of operating system (discussed in Chapter 14) was also not there during the era of first-generation computers. Hence, the first-generation computers were largely usable only by good electronics engineers who understood the logical structure of a computer in great detail and also knew machine and assembly language programming. The characteristic features of first-generation computers are as follows: 1. They were the fastest calculating devices of their time. 2. They were too bulky in size requiring large rooms for installation. 3. Thousands of vacuum tubes that were used emitted large amount of heat and burnt out frequently. Hence the rooms/areas in which these computers were located had to be properly airconditioned. 4. Each vacuum tube consumed about half a watt of power. Since a computer typically used more than ten thousand vacuum tubes, the power consumption of these computers was very high. 5. As vacuum tubes used filaments, they had a limited life. Since thousands of vacuum tubes were used in making one computer, these computers were prone to frequent hardware failures, their mean time between failures being as low as an hour. 6. Due to such low mean time between failures, these computers required almost constant maintenance. 7. In these computers, thousands of individual components had to be assembled manually by hand into functioning circuits. Hence commercial production of these computers was difficult and costly. 8. Since these computers were difficult to program and use, they had limited commercial use. Second Generation (1955-1964) A new electronic switching device called transistor [see Figure 1.2(b)] was invented at Bell Laboratories in 1947 by John Bardeen, Willian Shockley, and Walter Brattain. Transistors soon proved to be a much better electronic switching device than the vacuum tubes due to their following properties: 1. They were more rugged and easier to handle than tubes since they were made of germanium semiconductor material rather than glass. 2. They were highly reliable as compared to tubes since they had no part like a filament, which could burn out.
3. They could switch much faster (almost ten times faster) than tubes. Hence switching circuits made of transistors could operate much faster than their counterparts made of tubes. 4. They consume almost one-tenth the power consumed by a tube. 5. They were much smaller in size than a tube. 6. They were less expensive to produce. 7. They dissipated much less heat as compared to vacuum tubes. The second-generation computers were manufactured using transistors instead of vacuum tubes. Due to the properties of transistors listed above, the second-generation computers were more powerful, more reliable, less expensive, smaller, and cooler to operate than the first-generation computers. The second generation also experienced a change in storage technology. The memory of the second-generation computers was composed of magnetic cores. Magnetic cores were small rings made of ferrite which could be magnetized in either clockwise or anti-clockwise direction. Several magnetic cores were strung on a mesh of wires to construct large random access memories having storage capacity of few tens of kilo bytes. In 1957, magnetic tape was introduced as a faster and more convenient secondary storage medium. At that time, a single tape could hold the contents of approximately 1000 punched cards. Later magnetic disk storage was also developed, and magnetic disk and magnetic tape were the main secondary storage media used in second-generation computers, although punched" cards were still popular and widely used for preparing programs and data to be fed to the computer. On the software front, the second generation saw the emergence of high-level programming languages and batch operating systems. High-level programming languages like FORTRAN, COBOL, ALGOL and SNOBOL were developed during the second generation period, which were much easier for people to understand and work with than assembly or machine languages. Hence second-generation computers were easier to program and use than first-generation computers. The introduction of batch operating system allowed multiple jobs to be batched together and submitted at a time, and automatic transition from one job to another as soon as the former job finished. This concept helped in reducing human intervention while processing multiple jobs, resulting in faster processing, enhanced thoughput, and easier operation of second-generation computers. The second generation also saw an increase in the type of computer applications. The first-generation computers were mainly used for scientific computations. But in the second generation an increasing usage of computers was seen in business and industry for commercial data processing applications like payroll, inventory control, marketing, and production planning. The ease of use of second-generation computers also gave birth to the new profession of programmers and systems analysts in computing, which is more oriented towards usage
rather than design of computers. This also triggered the introduction of computer science related courses in several colleges and universities. The characteristic features of second-generation computers are as follows: 1. They were more than ten times faster than the first-generation computers. They were able to reduce computation times from milliseconds to microseconds. 2. They were much smaller in size than first-generation computers, requiring smaller space for installation. 3. Although the heat dissipation was much less than first-generation computers, still the rooms/areas in which the second-generation computers were located had to be properly airconditioned. 4. They consumed much less power than the first-generation computers. 5. They were much more reliable and less prone to hardware failures than the firstgeneration computers. 6. They had faster and larger primary and secondary storage devices as compared to first-generation computers. 7. They were much easier to program and use than the first-generation computers. Hence they had wider commercial use. 8. In these computers, thousands of individual transistors had to be assembled manually by hand into functioning circuits. Hence commercial production of these computers was difficult and costly. (a) A Vacuum Tube (b) A Transistor (c) An lC Chip Third Generation (1964-1975) Advances in electronics technology continued and in 1958, Jack St. Clair Kilby and Robert Noyce invented the first integrated circuit. Integrated circuits (called ICs) are circuits consisting of several electronic components like transistors, resistors, and capacitors grown on a single chip of silicon eliminating wired interconnection between components. The IC technology was also known as "microelectronics" technology because it made it possible to integrate larger number of circuit components into very small (less than 5 mm square) surface of silicon known as "chip" [see Figure 1.2(c)]. Initially, the integrated circuits contained only about ten to twenty components. This technology was named small scale integration (SSI). Later, with the advancement in technology for manufacturing ICs, it became possible to integrate up to about hundred components on a single chip. This technology came to be known as medium scale integration (MSI). The third generation was characterized by computers built using integrated circuits. The earlier ones used SSI technology and the later ones used MSI technology. ICs were much smaller, less expensive to produce, more rugged and reliable, faster in operation, dissipated less heat, and consumed much less power than circuits built by wiring electronic components manually. The net result was that third-generation computers were more powerful, more reliable, less expensive, smaller, and cooler to operate than the second-generation computers.
Parallel advancements in storage technologies allowed the construction of larger magnetic cores based random access memory, and larger capacity magnetic disks and magnetic tapes. Hence the third-generation computers typically had few megabytes (less than 5 Megabytes), of main memory and magnetic disks capable of storing few tens of megabytes of data per disk drive. On the software front, the third generation saw the emergence of standardization of highlevel programming languages, timesharing operating systems, unbundling of software from hardware, and the creation of an independent software industry. During the third generation, efforts were made to standardize some of the existing high-level programming languages so that programs written for one computer can be easily ported to and executed on another computer. This resulted in standardized FORTRAN and COBOL languages known as FORTRAN IV and COBOL 68, which were standardized by the American National Standards Institute (ANSI) in 1966 and 1968 repectively. They were also referred to as ANSI FORTRAN and ANSI COBOL. The idea was that as long as these standards are followed in program writing, a FORTRAN or a COBOL program could be run on any computer with an ANSI FORTRAN or ANSI COBOL compiler (see Chapter 12 for details). Additionally, some more high-level programming languages were introduced during the third-generation period. Notable among these were PL/1, PASCAL and BASIC. Unlike other high-level programming languages, which used compiler technology, BASIC was designed to use interpreter technology (see Chapter 12 for details). We saw that in second-generation computers batch operating system was used. In these systems, users had to prepare their data and programs and then submit them to the computer centre for processing. The operator at the computer centre collected these user jobs and fed them to the computer in batches at scheduled intervals. The output produced for each job was then sent to the computer centre counter for being returned to the respective users. The inevitable delay resulting from this batch processing approach was very frustrating to some users, especially programmers because often they had to wait for days to locate and correct a few program errors. To remedy this situation, John Kemeny and Thomas Kurtz of Dartmouth College introduced the concept of timesharing operating system. Timesharing operating system simultaneously allows a large number of users to directly access and share the computing resources in such a way that each user gets the illusion that no one else is using the computer. This is accomplished by having a large number of independent, relatively low-speed, online terminals simultaneously connected to the main computer. Each terminal can be used by a separate user providing him/her direct access to the main computer. Timesharing operating system allocates CPU time in such a way that all user programs have a brief share of the CPU time in turn. That is, each user program is allocated a very short period of CPU time (known as a time slice) one by one. The processing speed of the CPU allows the CPU to switch from one user job to another in rapid succession and to execute a small portion of each job in the allocated time slice until the job is completed. The processing speed of CPU is such that each user gets the illusion that he/she alone is using the computer. The introduction of timesharing concept helped in drastically improving the productivity of programmers, and also made
on-line systems feasible, resulting in new on-line applications like airline reservation systems, interactive query systems, etc. Till 1965, computer manufacturers sold their hardware along with all the associated software, and did not charge separately for the software they provided to customers. For example, buyers received language translators for all the languages that could run on the computers they purchased. From the user's standpoint, all this software was free. But the situation changed in 1969 when IBM and other computer manufacturers began to price their hardware and software products separately. This unbundling of software from hardware gave users an opportunity to invest only in software of their need and value. For example, now buyers could purchase only the language translators they needed and not all the language translators supported on the computers they purchased. This led to the creation of many new software houses and the beginning of an independent software industry. Another important concept introduced during the third-generation period was that of upward compatible family of computers. During this period, IBM introduced its System 360 which was conceived as a family of computers with upward compatibility. That is, the 360s were different sizes of mainframe systems based on the same machine language. This sharing of a single machine language enabled businesses to easily upgrade their computers without the usual costs of replacing peripheral equipment and modifying programs to run on new systems. That is, when a company outgrew one model it could easily move up to the next model without worrying about converting all its existing software. The development and introduction of minicomputers also took place during the thirdgeneration period. The computers built till the early 1960s were mainframe systems, which only very large companies could afford to purchase and use. Clearly, a need existed for low-cost smaller computers to fill the gaps left by the bigger, faster, and costlier mainframe systems. Several innovators recognized this need and formed new firms in the 1960s to produce smaller computers. The first commercially available minicomputer, the PDP-8 (Programmed Data Processor), was introduced in 1965 by Digital Equipment Corporation (DEC). It could easily fit in the corner of a room and did not require the attention of a full-time computer operator. It was based on timesharing operating system and could be accessed simultaneously by a number of users from different locations in the same building. Its cost was about one-fourth of the cost of a traditional mainframe system, making it possible for smaller companies to afford computers. It confirmed the tremendous demand for small computers for business and scientific applications, and by 1971 there were more than 25 computer manufacturers who had entered the minicomputer market. The characteristic features of third-generation computers are as follows: 1. They were much more powerful than the second-generation computers. They were capable of performing about l million instructions per second. 2. They were much smaller in size than second-generation computers, requiring smaller space for installation.
3. Although the heat dissipation was much less than second-generation computers, still the rooms/areas in which the third-generation computers were located had to be properly air conditioned. 4. They consumed much less power than the second-generation computers. 5. They were much more reliable and less prone to hardware failures than the secondgeneration computers. Hence the maintenance cost was much lower. 6. They had faster and larger primary memory and secondary storage devices as compared to second-generation computers. 7. They were totally general-purpose machines suitable for both scientific and commercial applications. 8. Their manufacturing did not require manual assembly of individual components into electronic circuits, resulting in reduced human labor and cost involved at assembly stage. Hence commercial production of these systems was easier and cheaper. However, highly sophisticated technology and expensive setup was required for the manufacture of IC chips. 9. Standardization of high-level programming languages allowed programs written for one computer to be easily ported to and executed on another computer. 10. Timesharing operating system allowed interactive usage and simultaneous use of these systems by a large number of users. 11. Timesharing operating system helped in drastically improving the productivity of programmers cutting down the time and cost of program development by several fold. 12. Timesharing operating system also made on-line systems feasible, resulting in the usage of these systems for new on-line applications. 13. Unbundling of software from hardware gave users of these systems an opportunity to invest only in software of their need and value. 14. The concept of upward compatible family of computers allowed the users of these systems to easily upgrade their system without worrying about converting all their existing software. 15. The minicomputers of the third-generation made computers affordable even by smaller companies. Fourth Generation (1975-1989) The average number of electronic components packed on a silicon chip doubled each year after 1965. This progress soon led to the era of large scale integration (LSI) when it was possible to integrate over 30,000 electronic components on a single chip, followed by very large scale integration (VLSI) when it was possible to integrate about one million electronic components on a single chip. This progress led to a dramatic development - the creation of a microprocessor. A microprocessor contains all the circuits needed to perform arithmetic logic and control functions, the core activities of all computers, on a single chip. Hence it became possible to build a complete computer with a microprocessor, a few additional primary storage chips, and other support circuitry. It started a new social revolution - the personal computer (PC) revolution. Overnight computers became incredibly compact. They became inexpensive 4o make and suddenly it became possible for anyone and everyone to own a computer. During the fourthgeneration period a large number of personal computers were deployed in both offices
and homes, which changed the work culture of most organizations. By 1978, the dominant personal computers were the Apple II from Apple Computer Inc. and the TRS80 model from the Radio Shack Division of Tandy Corporation. In 1980, IBM realized that the personal computer market was too promising to ignore and came out with its own PC in 1981, popularly known as IBM PC. Apple's and IBM's PCs became very popular during the fourth generation. Several other manufacturers used IBM's specification and designed their own PCs, which were popularly known as IBM compatible PCs or clones. The IBM PC and its clones became a popular standard for the PC industry during the fourth generation. During the fourth generation, magnetic core memories were replaced by semiconductor memories. This allowed computer manufacturers to provide large random access memories with very fast access time. Typical memory sizes were of the order of 16 Megabytes with access time of 200 nano seconds. On the other hand, hard disks became cheaper, smaller in size, and larger in capacity. Hence they became the standard in-built secondary storage device for all types of computer systems. Typical hard disk storage sizes in those days were of the order of few hundred of megabytes of data per disk drive. Parallely, in addition to magnetic tapes, floppy disks became very popular as a portable medium for porting programs and data from one computer system to another. In addition to the proliferation of PCs, during the fourth generation, there were significant advancements in the area of large-scale computer systems. In addition to improved processing and storage capabilities of mainframe systems, the fourth-generation saw the advent of supercomputers based on parallel vector processing and symmetric multiprocessing technologies. A supercomputer based on parallel vector processing technology contains a small number of custom-designed vector processors which are connected to a number of high-speed data access shared memory modules through a custom-designed, high-bandwidth crossbar switch network. On the other hand, a supercomputer based on symmetric multiprocessing technology uses commodity microprocessors, which are connected to a shared memory through a high-speed bus or a crossbar switch network. Primary builders of the supercomputers of the former category included Cray Research and ETA Systems, whereas of the latter category included IBM, Silicon Graphics and Digital Equipment Corporation. Another significant development during the fourth-generation period was the spread of high-speed computer networking, which enabled multiple computers to be connected together so that they could communicate and share data. Local area networks (LANs) became popular for connecting several dozen or even several hundred computers within an organization or within a campus, and wide area networks (WANs) became popular for connecting computers located at larger distances. This gave rise to network of computers and distributed systems. On the software front, there were several new developments that emerged to match the new technologies of the fourth generation. For example, several new operating systems were developed for PCs. Notable among these were MS-DOS and MS-Windows, which were, used on IBM PCs and its clones, and Apple's propriety OS, which was used on
Apple's PCs. Because PCs were to be used by individuals who were not computer professionals, to make computers more user friendly (easier to use), companies developed graphical user interfaces. A graphical user interface (GUI) provides icons (pictures) and menus (list of choices) that users can select with a mouse. This enables new computer users to learn to use a PC quickly. Several new PC-based applications were also developed to make the PCs a powerful tool. Notable among these were powerful word processing packages that allowed easy development of documents, spreadsheet packages that allowed easy manipulation and analysis of data organized in columns and rows, and graphics packages that allowed easy drawing of pictures and diagrams. Another very useful concept that became popular during the fourth-generation period was that of multiple windows on a single terminal screen. This feature allowed users to simultaneously see the current status of several applications in separate windows on the same terminal screen. In the area of software for large-scale, computers, the key technologies that became popular during the fourth-generation include multiprocessing operating systems, and concurrent programming languages. The multiprocessing operating systems allowed mainframe systems to be built of multiple processors (a main processor and several subordinate processors) in such a manner that the subordinate processors could be used to manage the user terminals and peripheral devices, allowing the main processor to concentrate on processing the main program. This helped in improving the overall performance of mainframe systems by several fold. The multiprocessing operating systems were also used with the supercomputers to extract the best performance from the large number of processors used in these systems. Concurrent programming languages further helped in effectively using the multiprocessing capabilities of these systems by allowing the programmers to write their applications in such a way that parts of the application can be executed in parallel on different processors. The most ambitious language of this type was ADA. During the fourth-generation period, the UNIX operating system also became very popular for use on large-scale systems. Additionally, due to the proliferation of computer networks, several new features were included in existing operating systems to allow multiple computers on the same network to communicate with each other and share resources. Some other software technologies that became popular during the fourth-generation period are C programming language, object-oriented software design, and object-oriented programming. C language combines the features of high-level programming languages with the efficiency of an assembly language. It allows the programmers to directly manipulate bits of data internal to the processor registers of the computer. As a result, C programs run significantly faster than programs written in other high-level programming languages, and C language can be effectively used for systems programming like design of compilers and operating systems. The primary objectives of object-oriented software design are to make programs generalized and to build software systems by combining reusable pieces of program codes called objects. To facilitate object-oriented software design, several object-oriented programming languages were introduced. Out of these,
C++ emerged as the most popular object-oriented language. It contains all the elements of the basic C language, but has been expanded to include numerous object-oriented programming features. The characteristic features of fourth-generation computers are as follows: 1. The PCs were much more smaller and cheaper than the mainframes or minicomputers of the third generation. 2. The mainframes and the supercomputers were much more powerful than the thirdgeneration systems. 3. Although the mainframes and the supercomputers required proper air conditioning of the rooms/areas in which they were located, no air conditioning was required for the PCs. 4. They consumed much less power than the third-generation computers. 5. They were much more reliable and less prone to hardware failures than the thirdgeneration computers. Hence the maintenance cost was negligible. 6. They had faster and larger primary memory and secondary storage devices as compared to third-generation computers. 7. They were totally general-purpose machines. 8. Their manufacturing did not require manual assembly of individual components into electronic circuits, resulting m reduced human labor and cost involved at assembly stage. Hence commercial production of these systems was easier and cheaper. However, highly sophisticated technology and expensive setup was required for the manufacture of LSI and VLSI chips. 9. Use of standard high-level programming languages allowed programs written for one computer to be easily ported to and executed on another computer. 10. Graphical user interface (GUI) enabled new computer users to learn how to use computers very quickly. 11. PC-based applications made the PCs a powerful tool for both office and home usage. 12. Multiprocessing coupled with concurrent programming languages allowed many new types of complex problems to be solved on the large-scale computers within reasonable time. 13. Network of computers enabled sharing of resources like disks, printers, etc. among multiple computers and their users. They also enabled several new types of applications involving interaction among computer users at geographically distant locations. Computer Supported Cooperative Working (CSCW), or groupware is one such application in which multiple members working on a single project and located at distant locations cooperate with each other using a network of computers. 14. Availability of C programming language allowed these systems to be effectively used for systems programming also like design of compilers and operating systems. 15. Object-oriented languages allowed development of reusable software modules, which helped in rapid software development. 16. In addition to unbundled software, these systems also used unbundled or add-on hardware feature which allowed users to invest only in the hardware configuration and software of their need and value. 17. The PCs of the fourth-generation made computers affordable even by individuals for their personal use at home.
Fifth Generation (1989-Present) The trend of further miniaturization of electronic components, the dramatic increase in the power of microprocessor chips, and the capacity of main memory and hard disk continued in the fifth generation. The VLSI technology became ULSI (Ultra Large Scale Integration) technology in the fifth generation resulting in the production of microprocessor chips having ten million electronic components. In fact, the speed of microprocessors and the size of main memory and hard disk doubled almost every eighteen months. The result was that many of the features found in the CPUs of large mainframe systems of the third and fourth generations became part of the microprocessor architecture in the fifth generation. This ultimately resulted in the availability of very powerful and compact computers becoming available at cheaper rates and the death of traditional large mainframe systems. Due to this fast pace of advancement in computer technology, we see more compact and more powerful computers being introduced almost every year at more or less the same price or even cheaper. Notable among these are portable notebook computers which give the power of a PC to their users even while travelling, powerful desktop PCs and workstations, powerful servers, and very powerful supercomputers (see Chapter 20 for a detailed description of these types of computers). Storage technology also advanced very fast making larger and larger main memory and disk storage available in newly introduced systems. Currently, PCs having 256 Megabytes (MB) of main memory and 4 to 20 Gigabytes (GB) of hard disk capacity are common. Similarly, workstations having 4 Gigabytes of main memory and few tens of Gigabytes of hard disk capacity are common. Larger hard disk space (of the order of few hundred Gigabytes) with better in-built reliability is now being supported with RAID (Redundant Array of Inexpensive Disks) technology that allows a bunch of disks to be configured as a single large disk. During the fifth generation, optical disks also emerged as a popular portable mass storage media. They are more commonly known as CD-ROM (Compact Disk - Read Only Memory) because they are mainly used for storing programs and data that are only read (not written/modified). In the area of large-scale systems, the fifth-generation period saw the emergence of more powerful supercomputers based on parallel processing technology. That is, the same multiprocessing concept of fourth generation was used to build shared memory and distributed memory parallel computers. In a shared memory parallel computer, a number of processors are connected to a common main memory by a high-speed bus or communication network, whereas in a distributed memory parallel computer, a number of processors, each with its own memory, are interconnected by a communication network. Parallel programming technique is used in these systems to break a problem into smaller problems and execute them parallely on the multiple processors of the system. While working on a problem, the various processors of a shared memory parallel computer use memory access mechanism for communication, whereas the various processors of a distributed memory parallel computer use message passing mechanism for communication. Distributed memory parallel computers have better scalability (can grow
larger in capability) than shared memory parallel computers, and are now also built by clustering together powerful commodity workstations by using a high-speed commodity switched network. This is also known as clustering technology. During the fifth-generation period there was tremendous outgrowth of computer networks. Communication technologies became faster day-by-day and more and more computers got networked together. This trend resulted in the emergence and popularity of the Internet and associated technologies and applications. The Internet made it possible for computer users sitting across the globe to communicate with each other within minutes by the use of electronic mail (known as e-mail) facility. A vast ocean of information became readily available to the computer users through the World Wide Web (known as WWW). And several new types of exciting applications like electronic commerce, virtual libraries, virtual classrooms, distance education, etc. emerged during the period. The tremendous processing power and the massive storage capacity of the fifthgeneration computers also made them a very useful and popular tool for a wide range of multimedia applications, which deal with information containing text, graphics, animation, audio, and video data. In general, the data size for multimedia information is much larger than plain text information because representation of graphics, animation, audio, or video media in digital form requires much larger number of bits than that required for representation of plain text. Because of this, multimedia computer systems require faster processor (for more quickly processing large amount of data), larger storage devices (for storing large data files), larger main memory (for running programs with large data size), good graphics terminals (for displaying graphics, animation and video), and input/output devices required to play any audio or video associated with a multimedia application program. The availability of multimedia computer systems resulted in a tremendous growth of multimedia applications during the fifth-generation period. In the area of operating systems, some of the concepts that gained popularity during the fifth-generation period include microkernels, multithreading, and distributed operating systems. Microkernel technology allows operating systems to be modeled and designed in a modular fashion. This makes operating systems easier to design and implement, easier to modify or add new services, and also allows those users who do not like a particular service provided by the operating system to implement and use their own service. Multhithreading technology is a popular way to improve application performance through parallelism. In traditional operating systems the basic unit of CPU utilization is a process, but in multithreading operating systems the basic unit of CPU utilization is a thread- In these operating systems, a process consists of an address space containing its instructions and data, and one or more thread sharing the same address space. Hence creation of a new thread, switching the CPU between threads, and sharing of resources between threads of the same process can be done more efficiently than between processes resulting in faster execution of the process and better overall system performance. A distributed operating system is an operating system for a set of computers networked
together whose goal is to make the multiple computers of the system appear as a single large virtual system to its users instead of a collection of distinct machines. Finally, in the area of programming languages, some of the concepts that gained popularity during the fifth generation are JAVA programming language, and parallel programming libraries like MPI (Message Passing Interface) and PVM (Parallel Virtual Machine). JAVA is a new programming language that is primarily used on the World Wide Web. It supports dynamic data types and dynamic protocols based on a concept called Java-based applets. This feature of Java allows web pages to have dynamic information and more interactivity with the users of web information. The MPI and PVM libraries have basically been introduced to standardize message passing parallel programs so that parallel programs developed for one message passing parallel computer can be easily ported and executed on other message passing parallel computers. The characteristic features of fifth-generation computers are as follows: 1. Portable PCs (called notebook computers) are much more smaller and handy than the PCs of the fourth generation allowing users to use computing facility even while travelling. 2. The desktop PCs and workstations are several times more powerful than the PCs of the fourth generation. 3. The supercomputers are several times more powerful than the mainframe systems and supercomputers of the fourth generation. 4. Although the supercomputers require proper airconditioning of the rooms/areas in which they are located, no aidconditioning is normally required for the notebook computers, desktop PCs, and workstations. 5. They consume much less power than their predecessors. 6. They are much more reliable and less prone to hardware failures than their predecessors. Hence the maintenance cost is negligible. 7. Many of the large-scale systems of fifth generation have hot-pluggable feature so that a failed component can be replaced with a new one without the need to shutdown the system, allowing the uptime of the system to be very high. 8. They have faster and larger primary memory and secondary storage devices as compared to their predecessors. 9. They are totally general-purpose machines. 10. Their manufacturing does not require manual assembly of individual components into electronic circuits, resulting in reduced human labor and cost involved at assembly stage. Hence commercial production of these systems is easier and cheaper. However, highly sophisticated technology and expensive setup (affordable only by a few organizations in the world) is required for the manufacture of ULSI chips. 11. Use of standard high-level programming languages and parallel programming libraries allows programs written for one computer to be easily ported to and executed on another computer. 12. More user-friendly interfaces with multimedia features make the systems easier to learn and use by anyone including small children. 13. Newer and more powerful applications, including multimedia applications, make the systems more useful in every walk of life.
14. More powerful supercomputers have allowed solving of many new types of complex problems, which were not possible to be solved before. 15. The explosion in the size of the Internet coupled with Internet-based tools and applications have made these systems influence the life of even common men and women. 16. The new operating systems concepts used in building these systems have further enhanced the performance and user-friendliness of these systems as compared to their predecessors. 17. These systems also use the concept of unbundled software and add-on hardware, allowing the users to invest only in the hardware configuration and software of their need and value. 18. With so many types of computers in all price ranges, today we have a computer for almost any type of user- whether the user is a small child or a scientist of world-fame. Figure 1.3 summarizes the key technologies and features of the various generations of computers discussed above. We have looked at the history of computing divided into five generations, and we have seen how quickly things have changed in the last few decades. But the rate of technological progress outlined in the preceding pages is not slowing down at all. As we enter into the 21st century, future generations of computers will evolve with higher capability and user friendliness. In fact, the fastest-growth period in the history of computing may be still ahead.18 1 Foundations of Computing Generation Key hardware technologies (Period) First (1942- Vacuum tubes; electromagnetic 1955) relay memory; punched cards secondary storage
Key software technologies
Third 1975)
Timesharing operating system; standardization of high-level programming languages; unbundling of software from hardware
Machine and assembly languages; stored program concept; mostly scientific applications Second (1955- Transistors; magnetic cores Batch operating system; 1964) memory; magnetic tapes and high-level programming disks secondary storage languages; scientific and commercial applications (1964- ICs with SSI and MSI technologies; larger magnetic cores memory; larger capacity disks and magnetic tapes secondary storage; minicomputers; upward compatible family of computers
Key characteristics
Bulky in size; highly unre limited commercial commercial production di and costly; difficult to use Faster, smaller, more reliabl easier to program than pre generation systems; comm production was still difficu costly Faster, smaller, more re easier and cheaper to pr commercially, easier to use easier to upgrade than pre generation systems; scie commercial and interactive o applications
Fourth (1975- ICs with VLSI technology; 1989) microprocessors; semiconductor memory; larger capacity hard disks as in-built secondary storage; magnetic tapes and floppy disks as portable storage media; personal computers; supercomputers based on parallel vector processing and symmetric multiprocessing technologies; spread of highspeed computer networks
Operating systems for PCs; GUI; multiple windows on a single terminal screen; multiprocessing operating systems; concurrent programming languages; UNIX operating system; C programming language; Object-oriented design and programming; PC-based applications; network-based applications; supercomputing applications
Small, affordable, reliable, easy to use PCs; more pow and reliable mainframe sy and supercomputers; totally g purpose machines; easie produce commercially; eas upgrade; rapid so development possible
Generation Key hardware technologies (Period) Fifth (1989- ICs with ULSI technology; Present) larger capacity main memory; larger capacity hard disks with RAID support; optical disks as portable read-only storage media; notebook computers; powerful desktop PCs and workstations; powerful servers; very powerful supercomputers; internet; cluster computing
Key software technologies Key characteristics Microkernal-based Portable computers; operating systems; powerful, cheaper, reliable, multithreading operating easier to use de systems; distributed machines; very pow operating systems; supercomputers; very high up parallel programming due to hot-pluggable compon libraries like MPI & PVM; totally general purpose mach JAVA; World Wide Web; easier to produce commerc multimedia applications; easier to upgrade; rapid soft internet-based development possible applications; more complex S supercomputing applications
Points to Remember 1. A computer is normally considered to be a calculative device that can perform arithmetic operations at enormous speed. More accurately, it is defined as a device that operates upon data. It is also known as a data processor since it not only computes in the usual sense but also performs other functions with the data. 2. The activity of processing data using a computer is called data processing. Data is the raw material used as input to data processing, and information is the processed data obtained as the output of data processing. Thus, information is data afrangedln an order and form that is useful to the people who receive it. 3. Computers are characterized by their being automatic, speed and accuracy of computing, diligence, versatility, power of remembering, and lack of intelligence and feelings. 4. The earliest device that qualifies as a digital computer is the "abacus" which was invented around 600 B.C.
5. Charles Babbage is considered to be the father of modern digital computers. 6. Some of the well known early computers are the MARK I (1937-44), the ATANASOFF-BERRY (1939-42), the ENIAC (1943-46), the ED VAC (1946-52), the EDS AC (1947-49), the MANCHESTER MARK I (1948), and the UNIVACI (1951). 7. Dr. John Von Neumann introduced the "stored program" concept, which considerably influenced the development of modern digital computers. Because of this feature, we often refer to modern digital computers as stored program digital computers. 8. "Generation" in computer talk is a step in technology. It provides a framework for the growth of the computer industry. Originally it was used to distinguish between various hardware technologies, but now it has been extended to include both the hardware and the software. 9. Till today, there are totally five computer generations, namely first, second, third, fourth and fifth generation. 10. The key hardware and software technologies and the key characteristics of the computers of the five different generations are summarized in Figure 1.3. Questions 1. What is a computer? Why is it also known as a data processor? 2. What is data processing? What are its basic activities? 3. Differentiate between data and information. Which is more useful to the people and why? 4. List out and explain some of the important characteristics of a computer. 5. What is meant by garbage-in-garbage-out? 6. Computers are automatic machines. Discuss. 7. Errors can occur in a computer, but these, are mainly due to human rather than technological weaknesses. Discuss. 8. Unlike human beings, a computer is free from monotony, tiredness, and lack of concentration. Discuss. 9. Versatility is one of the most wonderful things about the computer. Discuss. 10. Unlike human beings, a computer forgets or looses certain information only when it is asked to do so. Discuss. 11. A computer has no intelligence of its own. Discuss. 12. Computers are devoid of emotions. Discuss. 13. Which is the earliest device that qualifies as a digital computer? 14. What is an abacus? 15. Who is known as the father of modern digital computers and why? 16. Who invented the concept of stored program? Why is this concept so important? 17. Name the first computer that was designed on the stored program concept. 18. Why are modern digital computers often referred to as stored program digital computers? 19. Which was the first commercially produced computer? When and where was it first installed? 20. Give the full form of the following abbreviations used in computer terminology: IBM, ENIAC, EDVAC, EDSAC, UNIVAC. 21. What is meant by "generation" in computer terminology? How many computer generations are there till now?
22. List out the various computer generations along with the key characteristics of the computers of each generation. 23. List out the key hardware technologies used in building the computers of each of the five generations. 24. List out the key software technologies used in building the computers of each of the five generations. 25. What are the advantages of transistors over vacuum tubes? 26. What is an IC? How does it help in reducing the size of computers? 27. List out some of the advantages of IC technology over transistor technology. 28. Give the full form of the following abbreviations used in computer terminology: GIGO, IC, SSI, MSI, LSI, VLSI, ULSI. 29. Why were first and second generation computers more difficult and costlier to produce commercially than the computers of subsequent generations? 30. Name the technologies used for constructing main memory in the computers of first, second, third, and fourth generations. 31. Name the secondary storage media popularly used in the computers of first, second, third, fourth, and fifth generations. 32. Which key feature of third generation computers made them a highly productive tool for programmers and why? 33. Why was it not possible to take a program written for one computer and run it on some other computer in the first and second generation computers? 34. In what ways did the concept of unbundling of software from hardware affect the computer manufacturers and the users of computers? 35. What were the new types of computer applications which emerged during each of the following computer generations and what were the main driving technologies behind each: (a) Third generation (b) Fourth generation (c) Fifth generation 36. What is meant by upward compatible family of computers? How is this concept useful for computer users? 37. What is a microprocessor? How did it revolutionize the computer industry? 38. Name two new types of computers which became popular during the following computer generations: (a) Fourth generation (b) Fifth generation 39. Name some applications that emerged due to the emergence of computer networks. 40. What is hot-pluggable feature? How is it useful? 41. Write short notes on: (a) First generation computers (b) Second generation computers (c) Third generation computers (d) Fourth generation computers (e) Fifth generation computers 42. Name some of the representative computer systems of each of the five computer generations.
Chapter 2 Basic Computer Organization Even though the size, shape, performance, reliability, and cost of computers have been changing over the years, the basic logical structure (based on stored program concept) as proposed by Von Neumann has not changed. No matter what shape and size of computers we are talking about, all computer systems perform the following five basic operations for converting raw input data into information useful to their users: 1.
Inputting. The process of entering data and instructions into the computer system.
2. Storing. Saving data and instructions so that they are available for initial or for additional processing as and when required. 3. Processing. Performing arithmetic operations (add, subtract, multiply, divide, etc.) or logical operations (comparisons like equal to, less than, greater than, etc.) on data in order to convert them into useful information. 4. Outputting. The process of producing useful information or results for the user, such as a printed report or visual display. 5. Controlling. Directing the manner and sequence in which all of the above operations are performed. The goal of this chapter is to familiarize you with the computer system units that perform these functions. This chapter will provide you with an overview of computer systems as they are viewed by computer system architects. The internal architectural design of computers differs from one system model to another. However, the basic organization remains the same for all computer systems. A block diagram of the basic computer organization is shown in Figure 2.1. In this figure, the solid lines are used to indicate the flow of instruction and data, and the dotted lines represent the control exercised by the control unit. It displays the five major building blocks, or functional units of a digital computer system. These five units correspond to the five basic operations performed by all computer systems. The function of each of these units is described below. INPUT UNIT Data and instructions must enter the computer system before any computation can be performed on the supplied data. This task is performed by the input unit that links the external environment with the computer system. Data and instructions enter input units in forms that depend upon the particular device used. For example, data are entered from a keyboard in a manner similar to typing, and this differs from the way in which data are entered through a scanner, which is another type of input device. However, regardless of
the form in which they receive their inputs, all input devices must provide a computer with data that are transformed into the binary codes that the primary memory of a computer is designed to accept. This transformation is accomplished by units called input interfaces. Input interfaces are designed to match the unique physical or electrical characteristics of input devices to the requirements of the computer system. In short, the following functions are performed by an input unit: 1. It accepts (or reads) the list of instructions and data from the outside world. 2. It converts these instructions and data in computer acceptable form. 3. It supplies the converted instructions and data to the computer system for further processing. OUTPUT UNIT The job of an output unit is just the reverse of that of an input unit. It supplies the information obtained as a result of data processing to the outside world. Thus it links the computer with the external environment. As computers work with binary code, the results produced are also in the binary form. Hence, before supplying the results to the outside world, it must be converted to human acceptable (readable) form. This task is accomplished by units called output interfaces. Output interfaces are designed to match the unique physical or electrical characteristics of output devices (terminals, printers, etc.) to the requirements of the external environment. In short, the following functions are performed by an output unit: 1. It accepts the results produced by the computer, which are in coded form and hence cannot be easily understood by us. 2. It converts these coded results to human acceptable (readable) form. 3. It supplies the converted results to the outside world. The data and instructions that are entered into the computer system through input units have to be stored inside the computer before the actual processing starts. Similarly, the results produced by the computer after processing must also be kept somewhere inside the computer system before being passed on to the output units. Moreover, the intermediate results produced by the computer must also be preserved for ongoing processing. The storage unit of a computer system is designed to cater to all these needs. It provides space for storing data and instructions, space for intermediate results, and also space for the final results. In short, the specific functions of the storage unit are to hold (store): 1. All the data to be processed and the instructions required for processing (received from input devices). 2. Intermediate results of processing. 3. Final results of processing before these results are released to an output device. The storage unit of all computers is comprised of the following two types of storage: 1. Primary storage. The primary storage, also known as main memory, is used to hold pieces of program instructions and data, intermediate results of processing, and recently
produced results of processing of the job(s) that the computer system is currently working on. These pieces of information are represented electronically in the main memory chip's circuitry, and while it remains in the main memory, the central processing unit can access it directly at a very fast speed. However, the primary storage can hold information only while the computer system is on. As soon as the computer system is switched off or reset, the information held in the primary storage disappears. Additionally, the primary storage normally has limited storage capacity because it is very expensive. The primary storage of modern computer systems is made up of semiconductor devices. 2. Secondary storage. The secondary storage, also known as auxiliary storage, is used to take care of the limitations of the primary storage. That is, it is used to supplement the limited storage capacity and the volatile characteristic of primary storage. This is because secondary storage is much cheaper than primary storage and it can retain information even when the computer system is switched off or reset. The secondary storage is normally used to hold the program instructions, data, and information of those jobs on which the computer system is not working on currently, but needs to hold them for processing at a later time. The most commonly used secondary storage medium is the magnetic disk. ARITHMETIC LOGIC UNIT The arithmetic logic unit (ALU) of a computer system is the place where the actual execution of the instructions takes place during the processing operation. To be more precise, calculations are performed and all comparisons (decisions) are made in the ALU. The data and instructions, stored in the primary storage prior to processing, are transferred as and when needed to the ALU where processing takes place. No processing is done in the primary storage unit. Intermediate results generated in the ALU are temporarily transferred back to the primary storage until needed at a later time. Data may thus move from primary storage to ALU and back again to storage many times before the processing is over. After the completion of processing, the final results, which are stored in the storage unit, are released to an output device. The type and number of arithmetic and logic operations that a computer can perform is determined by the engineering design of the ALU. However, almost all ALUs are designed to perform the four basic arithmetic operations (add, subtract, multiply, and divide), and logic operations or comparisons such as less than, equal to, and greater than. CONTROL UNIT How does the input device know that it is time for it to feed data into the storage unit? How does the ALU know what should be done with the data once they are received? And how is it that only the final results are sent to the output device and not the intermediate results? All this is possible because of the control unit of the computer system. By selecting, interpreting, and seeing to the execution of the program instructions, the control unit is able to maintain order and direct the operation of the entire system. Although, it does not perform any actual processing on the data, the control unit acts as a
central nervous system for the other components of the computer system. It manages and coordinates the entire computer system. It obtains instructions from the program stored in main memory, interprets the instructions, and issues signals that cause other units of the system to execute them. CENTRAL PROCESSING UNIT The control unit and the arithmetic logic unit of a computer system are jointly known as the Central Processing Unit (CPU). The CPU is the brain of any computer system. In a human body, all major decisions are taken by the brain and the other parts of the body function as directed by the brain. Similarly, in a computer system, all major calculations and comparisons are made inside the CPU and the CPU is also responsible for activating and controlling the operations of other units of a computer system. THE SYSTEM CONCEPT You might have observed by now that we have been referring to a computer as a system (computer system). What can be the reason behind this? To know the answer let us first consider the definition of a system. A system is a group of integrated parts that have the common purpose of achieving some objective(s). So, the following three characteristics ape key tea system: 1. A system has more than one element. 2. All the elements of a system are logically related. 3. All the elements of a system are controlled in such a way that the system goal is achieved. Since a computer is made up of integrated components (input, output, storage, and CPU) that work together to perform the steps called for in the program being executed, it is a system. The input or output units cannot function until they receive signals from the CPU. Similarly, the storage unit or the CPU alone is of no use. So the usefulness of each unit depends on other units and can be realized only when all units are put together (integrated) to form a system. Points to Remember 1. All computer systems perform the following five basic operations for converting raw input data into useful information - inputting, storing, processing, outputting, and controlling. 2. The main components of any computer system are shown in Figure 2.1. 3. The input unit allows data and instructions to be fed to the computer system from the outside world in computer acceptable form. 4. The input interfaces transform the input data and instructions fed to the computer through its input devices into the binary codes acceptable to the computer.
5. The output unit allows the computer system to supply the information obtained as a result of data processing to the outside world in human acceptable (readable) form. 6. The output interfaces transform the information obtained as a result of data processing from binary form to human acceptable (readable) form. 7. The storage unit of a computer system holds the data and instructions to be processed, and the intermediate and final results of processing. 8. The storage unit of all computers is comprised to two types of storage - primary and secondary storage. The primary storage is used to hold pieces of program instructions and data, intermediate results, and recently produced results of processing of the job(s) that the computer system is currently working on. On the other hand, the secondary storage is normally used to hold the program instructions, data, and information of those jobs on which the computer system is not working on currently but needs to hold them for use at a later time. 9. As compared to primary storage, the secondary storage is slower in operation, larger in capacity, cheaper in price, and can retain information even when the computer system is switched off or reset. 10. During data processing, the actual execution of the instructions takes place in the Arithmetic Logic Unit (ALU) of a computer system. 11. The control unit of a computer system manages and coordinates the operations of all the other components of the computer system. 12. The control unit and the arithmetic logic unit of a computer system are jointly known as the Central Processing Unit (CPU), which serves as the brain of the computer system and is responsible for controlling the operations of all other units of the system. 13. A computer is often referred to as a computer system because it is made up of integrated components (input, output, storage, and CPU) that work together to perform the steps called for in the program being executed. Questions 1. What are the five basic operations performed by any computer system? 2. Draw a block diagram to illustrate the basic organization of a computer system and explain the functions of the various units. 3. What is an input interface? How does it differ from an output interface? 4. How many types of storage are normally there in the storage unit of a computer system? Justify the need for each storage type. 5. Differentiate between the characteristics of the primary storage and the secondary storage of a computer system. 6. What are the basic components of the CPU of a computer system? Describe the roles of each of the components in the functioning of a computer system. -_ 7. List out the logical steps taken by a computer system along with the roles of the main units of the computer system in each step while transforming an input data to useful information for presentation to the user. 8. What is a system? Why is a computer often referred to as a computer system?
Chapter 3 Number Systems We have already seen in the previous chapter that inside a computer system, data is stored in a format that cannot be easily read by human beings. This is the reason why input and output (I/O) interfaces are required. Every computer stores numbers, letters, and other special characters in a coded form. Before going into the details of these codes, it is essential to have a basic understanding of the number system. So the goal of this chapter is to familiarize you with the basic fundamentals of number system. It also introduces some of the commonly used number systems by computer professionals and the relationship among them. NON-POSITIONAL NUMBER SYSTEMS Number systems are basically of two types: non-positional and positional. In early days, human beings counted on fingers. When ten fingers were not adequate, stones, pebbles, or sticks were used to indicate values. This method of counting uses an additive approach or the non-positional number system. In this system, we have symbols such as I for 1, II for 2, III for 3, IIII for 4, IIIII for 5, etc. Each symbol represents the same value regardless of its position in the number and the symbols are simply added to find out the value of a particular number. Since it is very difficult to perform an arithmetic with such a number system, positional number systems were developed as the centuries passed. POSITIONAL NUMBER SYSTEMS In a positional number system, there are only a few symbols called digits, and these symbols represent different values depending on the position they occupy in the number. The value of each digit in such a number is determined by three considerations: 1. The digit itself, 2. The position of the digit in the number, and 3. The base of the number system (where base is defined as the total number of digits available in the number system). The number system that we use in our day-to-day life is called the decimal number system. In this system, the base is equal to 10 because there are altogether ten symbols or digits (0, 1, 2, 3, 4, 5, 6, 7, 8, 9). You know that in the decimal system, the successive positions to the left of the decimal point represent units, tens, hundreds, thousands, etc. But you may not have given much attention to the fact that each position represents a specific power of the base (10). For example, the decimal number 2586 (written as 258610) consists of the digit 6 in the units position, 8 in the tens position, 5 in the hundreds position, and 2 in the thousands position, and its value can be written as: (2x1000)+ (5x100)+ (8x10)+ (6xl), or 2000 + 500 + 80 + 6, or 2586
It may also be observed that the same digit signifies different values depending on the position it occupies in the number. For example, In 2586io the digit 6 signifies 6 x 10° = 6 In 2568io the digit 6 signifies 6 x 10 1 = 60 In 2658,o the digit 6 signifies 6 x 102 = 600 In 6258io the digit 6 signifies 6 x 103 = 6000 Thus any number can be represented by using the available digits and arranging them in various positions. The principles that apply to the decimal number system also apply to any other positional number system. It is important only to keep track of the base of the number system in which we are working. The following two characteristics are suggested by the value of the base in all positional number systems: 1. The value of the base determines the total number of different symbols or digits available in the number system. The first of these choices is always zero. 2. The maximum value of a single digit is always equal to one less than the value of the base. Some of the positional number systems commonly used in computer design and by computer professionals are discussed below. Binary Number System The binary number system is exactly like the decimal number system except that the base is 2 instead of 10. We have only two symbols or digits (0 and 1) that can be used in this number system. Note that the largest single digit is 1 (one less than the base). Again, each position in a binary number represents a power of the base (2). As such, in this system, the rightmost position is the units (2°) position, the second position from the right is the 2's (21) position and proceeding in this way we have 4's (2 2) position, 8's (23) position, 16's (24) position, and so on. Thus, the decimal equivalent of the binary number 10101 (written as IOIOI2) is: (1 x 24) + (0 x 23) + (1 x 22) + (0 x 21) + (1 x 2°), or 16 + 0 + 4 + 0+1, or 21 In order to be specific about which system we are referring to, it is common practice to indicate the base as a subscript. Thus we write: 101012 = 2110 "Binary digit" is often referred to by the common abbreviation bit. Thus, a "bit" in computer terminology means either a 0 or a 1. A binary number consisting of 'n' bits is called an n-bit number. Figure 3.1 lists all the 3-bit numbers along with their decimal
equivalent. Remember that we have only two digits, 0 and 1, in the binary system, and hence the binary equivalent of the decimal number 2 has to be stated as 10 (read as one, zero). Another important point to note is that with 3 bits (positions), only 8 (23) different patterns of 0s and Is are possible and from Figure 3.1 it may be seen that a 3-bit number can have one of the 8 values in the range 0 to 7. In fact, it can be shown that any decimal number in the range 0 to 2""1 can be represented in the binary form as an n-bit number. I Biliary Decimal Equivalent 000 001 010 011 100 101 110 111
0 1 2 3 4 5 6 7
Figure 3.1. 3-bit numbers with their decimal values. Every computer stores numbers, letters, and other special characters in binary form. There are several occasions when computer professionals have to know the raw data contained in a computer's memory. A common way of looking at the contents of a computer's memory is to print out the memory contents on a line printer. This printout is called a memory dump. If memory dumps were to be printed using binary numbers, the computer professionals would be confronted with many pages of 0s and Is. Working with these numbers would be very difficult and error prone. Because of the quantity of printout that would be required in a memory dump of binary digits and the lack of digit variety (0s and Is only), two number systems, octal and hexadecimal, are used as shortcut notations for binary. These number systems and their relationship with the binary number system are explained below. Octal Number System In the octal number system the base is 8. So in this system there are only eight symbols or digits: 0, 1,2, 3, 4, 5, 6, and 7 (8 and 9 do not exist in this system). Here also the largest single digit is 7 (one less than the base). Again, each position in an octal number represents a power of the base (8). Thus the decimal equivalent of the octal number 2057 (written as 20578) is: (2 x 83) + (0 x 82) + (5 x 81) + (7 x 8°), or 1024 + 0 + 40 + 7, or 1071 Hence, 20578= 107110
Observe that since there are only 8 digits in the octal number system, 3 bits (2 3 = 8) are sufficient to represent any octal number in binary (see Figure 3.1). Hexadecimal Number System The hexadecimal number system is one with a base of 16. The base of 16 suggests choices of 16 single-character digits or symbols. The first 10 digits are the digits of the decimal number system - 0, 1, 2, 3, 4, 5, 6, 7, 8, 9. The remaining six digits are denoted by the symbols A, B, C, D, E, and F representing the decimal values 10, 11, 12, 13, 14, and 15 respectively. In the hexadecimal number system, therefore, the letters A through F are number digits. The number A has a decimal equivalent value of 10 and the number F has a decimal equivalent value of 15. Thus, the largest single digit is F or 15 (one less than the base). Again, each position in the hexadecimal system represents a power of the base (16). Thus the decimal equivalent of the hexadecimal number 1AF (written as 1AFI6) is: (1 x 162) + (A x 161) + (F x 16°), or (1x256)+ (10x16) 256+160+ 15, or 431 (15xl), or Hence, 1AFI6 = 43110 Observe that since there are only 16 digits in the hexadecimal number system, 4 bits (24 =16) are sufficient to represent any hexadecimal number in binary. CONVERTING FROM ONE NUMBER SYSTEM TO ANOTHER Numbers expressed in decimal number system are much more meaningful to us than are values expressed in any other number system. This is mainly because of the fact that we have been using decimal numbers in our day-today life right from childhood. However, any number value in one number system can be represented in any other number system. Because the input and the final output values are to be in decimal, computer professionals are often required to convert numbers in other number systems to decimal and vice-versa. There are many methods or techniques that can be used to convert numbers from one base to another. We will see one technique used in converting to base 10 from any other base and a second technique to be used in converting from base 10 to any other base. Converting to Decimal from Another Base The following three steps are used to convert to a base 10 value from any other number system: Step 1: Determine the column (positional) value of each digit (this depends on the position of the digit and the base of the number system). Step 2: Multiply the obtained column values (in Step 1) by the digits in the corresponding columns. Step 3: Sum the products calculated in Step 2. The total is the equivalent value in decimal.
Example 3.1. 11OO12 = ?IO
Example 3.2. 47068 = ?,o
Solution:
Solution:
Step 1: Determine
column
values
Column Number (from Column Value right) 1 2345
Step 1: Determine
Column Number (from right) Column Value
2° = 2X- = 1 = 2 = 4 = 1 22 = 23 = 8 = 16 234 4 2 =
Step 2: Multiply column values by corresponding column digits 16 8 4 2 1 xl xl xO xO xl 16 8 0 0 1 Step 3: Sum the products 16 + 8 + 0 + 0+1 =25 Hence, 110012 = 2510 Step 2: Multiply column values by corresponding column digits 512 64 8 1 x4 x7 xO x6 2048 448 0 6 Step 3: Sum the products 2048 + 448 + 0 + 6 = 2502 Hence, 47068 = 250210 Solution: 1AC16 = 1 x 162 + A x 161 + C x 16° = 1x256+10x16+12x1 = 256+ 160+ 12 = 42810 Example 3.4. 40527 = ?, Solution: 40527 = 4x73 + 0x72 + 5x7' + 2x7° = 4x 343+ 0x 49+ 5x7 + 2x1 = 1372 + 0 + 35+2 = 140910
Column values
8° = 82 = = 1 = 8 83 = = 64 = 512
Example 3.5. 40526 = ?10 Solution: 40526 = 4x63 + 0x62 + 5x61 + 2x6° = 4x216 + 0x36 + 5x6 + 2x1 = 864 + 0 + 30 + 2 = 896,0 . Comparing this result with the result obtained in Example 3.4, we find that although the digits (4052) are same for both the numbers, but their decimal equivalent is different. This is because of the fact that the number in Example 3.4 is represented in base 7 number system whereas the number in Example 3.5 is represented in base 6 number system. Example 3.6. 110014 = ?10 Solution: Solution: 110014 =lx44+lx43 + Ox42 + Ox41 + 1x4° = 1 x 256 + 1 x 64 + 0x16 + 0x4 + lxl = 256 + 64 + 0 + 0+1 = 32110 Compare the result with Example 3.1. Example 3.7. IAC13 = ?10 IAC13= 1 x 132 + A x 131 + C x 13U = 1 x 169+ 10x13 + 12 x 1 = 31110 Compare the result with Example 3.3. Converting from Decimal to Another Base (Division-Remainder Technique) The following four steps are used to convert a number from decimal to another base: Step 1: Divide the decimal number to be converted by the value of the new base. Step 2: Record the remainder from Step 1 as the rightmost digit (least significant digit) of the new base number. Step 3: Divide the quotient of the previous divide by the new base. Step 4: Record the remainder from Step 3 as the next digit (to the left) of the new base number.
Repeat Steps 3 and 4, recording remainders from right to left, until the quotient becomes zero in Step 3. Note that the last remainder thus obtained will be the most significant digit (MSD) of the new base number. Example 3.8. 2510 = ?2 Solution: Steps 1 and 2: Steps 3 and 4: Steps 3 and 4: Steps 3 and 4: Steps 3 and 4:
25/2 = 12 ahd remainder 1 12/2= 6 and remainder 0 6/2 = 3 and remainder 0 3/2 = 1 and remainder 1 1/2= 0 and remainder 1
As mentioned in Steps 2 and 4, the remainders have to be arranged in the reverse order so that the first remainder becomes the least significant digit (LSD) and the last remainder becomes the most significant digit (MSD). Hence, 2510= 110012 Compare the result with Example 3.1. Converting from a Base Other Than 10 to a Base Other Than 10 The following two steps are used to convert a number from a base other than 10 to a base other than 10: Step 1: Convert the original number to a decimal number (base 10). Step 2: Convert the decimal number so obtained to the new base number. Example 3.15. 5455 = ?4 Solution: Step 1: Convert from base 6 to base 10 545 = 5 x 62 + 4 x 61 + 5 x 6° =5x36+4x6+5x1 = 180 + 24 + 5 = 20910 Example 3.16 illustrates the method of converting a number from binary to octal. Similarly, Example 3.17 shows how to convert a number from binary to hexadecimal. However, these are lengthy procedures and shortcut methods can be used when we desire such conversions. These shortcut methods are described below.
Shortcut Method for Binary to Octal Conversion The following steps are used in this method: Step 1: Divide the binary digits into groups of three (starting from the right). Step 2: Convert each group of three binary digits to one octal digit. Since there are only 8 digits (0 to 7) in the octal number system (refer to Figure 3.1), 3 bits (2 3 = 8) are sufficient to represent any octal number in binary. Since decimal digits 0 to 7 are equal to octal digits 0 to 7, binary to decimal conversion can be used in this step. Example 3.18. 1011102 = ?8 Solution: Step 1: Divide the binary digits into groups of 3 starting from right (LSD). 101 110 Step 2: Convert each group into one digit of octal (use binary-to-decimal conversion). 1012=lx22 + 0x21 + lx2° =4+0+1 = 58 1102= Ix22+1 x2'+0x2° =4+2+0 = 68 Hence, 1011102 = 568 Compare the result with the result of Example 3.16. Example 3.19. 11010102 = ?8 Solution: 1101010a =001 101 010 (Group of 3 digits from right) = 1528 (Convert each group to an octal digit) Hence, 11010102 = 1528 Shortcut Method for Octal to Binary Conversion The following steps are used in this method: Step 1: Convert each octal digit to a 3 digit binary number (the octal digits may be treated as decimal for this conversion). Step 2: Combine all the resulting binary groups (of 3 digits each) into a single binary number. Example 3.20. 5628 = ?2 Solution: Step 1: Convert each octal digit to 3 binary digits. 58=1012 68 =1102 28
= 0102 Step 2: Combine the binary groups. 5628 = 101/5 110/6 010/2 Hence, 5628 = 1011100102 Shortcut Method for Binary to Hexadecimal Conversion The following steps are used in this method: Step 1: Divide the binary digits into groups of four (starting from the right). Step 2: Convert each group of four binary digits to one hexadecimal digit. Remember that hexadecimal digits 0 to 9 are equal to decimal digits 0 to 9, and hexadecimal digits A to F are equal to decimal values 10* to 15. "Hence for this step, the binary to decimal conversion procedure can be used, but the decimal values 10 to 15 must be represented as hexadecimal A to F. Example 3.23. 101101011002 = ?16 Solution: Step 1 : Divide the binary digits into groups of 4. 1101 0011 Step 2: Convert each group of 4 binary digits to 1 hexadecimal digit. 11012 -=-1 x 23 + 1 x 22 + 0 x 21 + 1 x 2° =8+4+0+1 = 1310 = D16 00112 = 0 x 23 + 0 x 22 + 1 x 21 + 1 x 2° =0+0+2+1 = 316 Hence, 110100112 = D3,6 Compare the result with the result of Example 3.17. Shortcut Method for Hexadecimal to Binary Conversion The following steps are used in this method: Step 1: Convert the decimal equivalent of each hexadecimal digit to 4 binary digits. Step 2: Combine all the resulting binary groups (of 4 digits each) into a single binary number. Example 3.24. 2AB16 = ?2 Solution: Step 1: Convert the decimal equivalent of each hexadecimal digit to 4 binary digits. 216 = 210 = 00102
A16=10I0 =10102 B16=ll10 =l0ll2 Step 2: Combine the binary groups. 2AB16 = 0010/2 1010/A 1011/B Hence, 2AB16 = 0010101010112 Figure 3.2 summarizes the relationship between the decimal, binary, hexadecimal, and octal number systems. Note that the maximum value for a single digit of octal (7) is equal to the maximum value of three digits of binary. The value range of one digit of octal duplicates the value range of three digits of binary. If octal digits are substituted for binary digits, the substitution is on a one-to-three basis. Thus, computers that print octal numbers instead of binary, while taking memory dump, save one-third of the printing space and time. Decimal Hexadecimal Binary Octal 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0 1 2 3 4 5 6 7 8 9 A B C D E F 10
0 1 10 11 100 101 110 111 1000 1001 1010 1011 1100 1101 1110 Mi 10000
0 1 2 3 4 5 6 7 10
Figure 3.2. Relationship among Decimal, Binary, Hexadecimal and Octal number systems. Similarly, note that the maximum value of one digit in hexadecimal is equal to the maximum value of four digits in binary. Thus, the value range of one digit of hexadecimal is equivalent to the value range of four digits of binary. Therefore, hexadecimal shortcut notation is a one-to-four reduction in the space and time required for memory dump. FRACTIONAL NUMBERS
In binary number system, fractional numbers are formed in the same general way as in the decimal number system. For example, in the decimal number system, 0.235 = (2 x 10"') + (3 x 10"2) + (5 x 10"3) and 68.53 = (6 x 10') + (8x 10°) +(5 + (3 x 10"2) Similarly, in the binary number system, 0.101 =(1 x2"') + (0x2"2) 10.01 =(1 x21) + <0x2°) + x2"3)and 10.01 = (1x21) + (0x20) + 0x1-1) + (1x2-2) Thus, the binary point serves the same purpose as the decimal point. Some of the positional values in the binary number system are given below. Position 4 -4 Position Value 24 2-4 Quantity 16 1/16 Represented
3
2
1
0
Binary Point -1
-2
-3
23
22
21
20
Binary Point 2-1
2-2
2-3
8
4
2
1
Binary Point ½
¼
1/8
In general, a number in a number system with base b would be written as: an an.i... ao. a.i a_2... a.m and would be interpreted to mean an x bn + an., x b"'1 + ... + ao x b° + a4 x b"1 + a.2 x b"2 + .;. + a.m x b"m The symbols an, an.h ..., a.m used in the above representation should be one of the b symbols allowed in the number system. Thus, as per the above mentioned general rule, 46.328 = (4 x 81) + (6 x 8°) + (3 x 8'1) + (2 x 8"2) and 5A.3C16 = (5 x 161) + (A x 16°) + (3 x 16"1) + (C x 16"2) Example 3.26. Find the decimal equivalent of the binary number 110.101 Solution: 110.1012 = 1 x22+ 1 x2' + 0x2° + 1 x2" 0 x 2"1 + 0x2-2 + 1 x 2-3 = 4 + 2 + 0 + 0.5 + 0 + 0.125 = 6 + 0.5 + 0.125 = 6.62510 Points to Remember 1.
Number systems are basically of two types: non-positional and positional.
2. In a non-positional number system, each symbol represents the same value regardless of its position in the number and the symbols are simply added to find out the value of a particular number. It is very difficult to perform arithmetic with such a number system. 3. In a positional number system, there are only a few symbols called digits, and these symbols represent different values depending on the position they occupy in the number. The value of each digit in such a number is determined by three considerations. • The digit itself, • The position of the digit in the number, and • The base of the number system (where base is defined as the total number of digits available in the number system). 4. The number system that we use in our day-to-day life is called the decimal number system. In this system, the base is equal to 10 because there are altogether ten symbols or digits (0, 1, 2, 3, 4, 5, 6, 7, 8, 9). 5. Some of the positional number systems commonly used in computer design and by computer professionals are binary (for which base is 2), octal (for which base is 8), and hexadecimal (for which base is 16). 6. The relationship among decimal, binary, hexadecimal and octal number systems are given in Figure 3.2.44 I Foundations of Computing 7. Readers must know the techniques for the following types of conversions to convert numbers from one base to another: • Converting to decimal from another base • Converting from decimal to another base • Converting from a base other than 10 to a base other than 10 • Shortcut method for binary to octal conversion • Shortcut method for octal to binary conversion • Shortcut method for binary to hexadecimal conversion • Shortcut method for hexadecimal to binary conversion 8. Fractional numbers are formed in the same way as in the decimal number system. In general, a number in a number system with base b would be written as: anan.1... ao. a.2... a.m and would be interpreted to mean an x b" + an., x b""1 + ... + ao x b° + a.! x b"1 + a.2 x b"2 + ... + a.m x b'm The symbols an, an.i, ..., a.m used in the above representation should be one of the b symbols allowed in the number system. Questions 1. What is the difference between positional and non-positional number systems? Give examples of both types of number systems. 2. What is meant by the base of a number system? Give examples to illustrate the role of base in positional number systems. 3. What is the value of the base for decimal, hexadecimal, binary and octal number systems? 4. Give an example for octal number system to show that the same digit may signify different values depending on the position it occupies in the number.
5. What will be the total number of different symbols or digits and the maximum value of a single digit for the following number systems: (a) Number system with base 5 (b) Number system with base 20 (c) Number system with base 9 (d) Number system with base 12 6. What is a 'bit' in computer terminology? How many different patterns of bits are possible with (a) 6 bits (b) 7 bits (c) 8 bits 7. Explain the meaning of the term "memory dump". 8. Why are octal and/or hexadecimal number systems used as shortcut notation? 9. Find out the decimal equivalent of the following binary numbers: (a) 1101011 (f) 1000 (b) 11010 (g) 10110001100 (c) 10110011 (h) 1010101100 (d) 11011101 (i) 110001 (e) 1110101 (j) 111 10. Find out the octal equivalent of the binary numbers of Question 9. 11. Find out the hexadecimal equivalent of the binary numbers of Question 9. 12. Convert the following numbers to decimal numbers: (a) 1101102 (c) 2A3B16 (b) 25736 (d) 12349 13. Convert the following decimal numbers to binary numbers: (a) 43510 (c) 3210 (b) 169410 (d) 135,o 14. Convert the decimal numbers of Question 13 to octal numbers. 15. Convert the decimal numbers of Question 13 to hexadecimal numbers. 16. Carry out the following conversions: (a) 1256 = ?4 (b) 249 = ?3 (c) ABC16 = ?8 17. Convert the following numbers to their binary equivalent: (a) 2AC16 (c) 26148 (b) FAB16 (d) 5628 18. Find the decimal equivalent of the following numbers: (a) lll.Oh (c) 247.658 (b) 1001.0112 (d) A2B.D416
Chapter 4 Computer Codes In the previous chapter we have discussed about true or "pure" binary numbers. In this chapter we will see how these binary numbers are coded to represent characters in computers. Thus the goal of this chapter is to present the formats used in computers to record data. Although many coding schemes have been developed over the years, we will be discussing only the most commonly used computer codes. Numeric data is not the only form of data that is to be handled by a computer. We often require to process alphanumeric data also. An alphanumeric data is a string of symbols where a symbol may be one of the letters A, B, C, ..., Z or one of the digits 0, 1,2, ..., 9 or a special character, such as+ -*/,.() = (space or blank) etc. An alphabetic data consists of only the letters A, B, C, ..., Z and the blank character. Similarly, numeric data consists of only numbers 0, 1,2, ..., 9. However, any data must be represented internally by the bits 0 and 1. As such, binary coding schemes are used in computers to represent data internally. In binary coding, every symbol that appears in the data is represented by a group of bits. The group of bits used to represent a symbol is called a byte. To indicate the number of bits in a group, sometimes a byte is referred to as "n-bit byte" where the group contains n bits. However, the term byte is commonly used to mean an 8-bit byte (a group of 8 bits) because most of the modern computers use 8 bits to represent a symbol. BCD CODE The Binary Coded Decimal (BCD) code is one of the early memory codes. It is based on the idea of converting each digit of a decimal number into its binary equivalent rather than converting the entire decimal value into a pure binary form. This facilitates the conversion process to a great extent. The BCD equivalent of each decimal digit is shown in Figure 4.1. Since 8 and 9 require 4 bits, all decimal digits are represented in BCD by 4 bits. You have seen in Example 3.9 that 42i0 is equal to IOIOIO2 in a pure binary form. Converting 42i0 into BCD, however, produces the following result: 4210 = 0100/4 0010/2 or 01000010 in BCD Decimal Digits BCD Equivalent 0 1 2 3 4 5 6 7 8 9
0000 0001 0010 0011 0100 0101 0110 0111 1000 1001
Figure 4.1. BCD equivalent of decimal digits. Note that each decimal digit is independently converted to a 4 bit binary number and hence, the conversion process is very easy. Also note that when 4 bits are used, altogether 16 (24) configurations are possible (refer to hexadecimal number system). But from Figure 4.1 you can see that only the first 10 of these combinations are used to represent decimal digits. The remaining 6 arrangements (1010, 1011, 1100, 1101, 1110 and 1111) have decimal values from 10 to 15. These arrangements are not used in BCD coding. That is, 1010 does not represent lO,o in BCD. Instead, In the above discussion, we have used a group of 4 bits to represent a digit (character) in BCD. 4-bit BCD coding system can be used to represent only decimal numbers because 4 bits are insufficient to represent the various characters used by a computer. Instead of using 4 bits with only 16 possible characters, computer designers commonly use 6 bits to represent characters in BCD code. In the 6-bit BCD code, the four BCD numeric place positions are retained, but two additional zone positions are added. With 6 bits, it is possible to represent 64 (26) different characters. This is a sufficient number to code the decimal digits (10), alphabetic letters (26), and other special characters (28). Figure 4.2 illustrates the coding of alphabetic and numeric characters in BCD. In Chapter 3 we have seen the use of octal and hexadecimal number systems as shortcut notations for binary. Because BCD is a 6-bit code, it can be easily divided into two 3-bit groups. Each of these 3-bit groups can be represented by 1 octal digit. Thus octal number system is used as shortcut notation for memory dump by computers that use BCD code for internal representation of characters. This results in a one-to-three reductions in the volume of memory dump. Figure 4.2 also shows the octal equivalent of the alphabetic and numerical characters coded in BCD.48 I Foundations of Computing Character BCD Code Octal Equivalent A B C D E F G H I
Zone 11 11 11 11 11 11 11 11 11
Digit 0001 0010 0011 0100 0101 0110 0111 1000 1001
61 62 63 64 65 66 67 70 71
J K L M N
10 10 10 10 10
0001 0010 0011 0100 0101
41 42 43 44 45
O P Q R
10 10 10 ' 10
0110 0111 1000 1001
46 47 50 51
S T U V w X Y z
01 01 01 01 01 01 01 01
0010 0011 0100 0101 0110 0111 1000 1001
22 23 24 25 26 27 30 31
1 00 0001 01 2 00 0010 02 3 00 0011 03 4 00 0100 04 5 00 0101 05 6 00 0110 06 7 00 0111 07 8 00 1000 10 9 00 1001 11 0 00 1010 12 Figure 4.2. Alphabetic and numeric characters in BCD along with their Octal equivalent.Chapter 4 ■ Computer Codes Example 4.1. Show the binary digits used to record the word BASE in BCD. Solution: B = 110010 in BCD binary notation A = 110001 in BCD binary notation S = 010010 in BCD binary notation E = 110101 in BCD binary notation So the binary digits 110010/B 110001/A 010010/S 110101/E will record the word BASE in BCD. Example 4.2. Using octal notation, show the BCD coding for the word DIGIT. Solution:
D = 64 in BCD octal notation I = 71 in BCD octal notation G = 67 in BCD octal notation I = 71 in BCD octal notation T = 23 in BCD octal notation So the BCD coding for the word DIGIT in octal notation will be 64/D 21/I 67/G 21/I 23/T EBCDIC The major problem with BCD code is that only 64 (26) different characters can be represented in it. This is not sufficient for providing decimal numbers (10), lower-case letters (26), capital letters (26), and a fairly large number of other special characters (28+). Hence, the BCD code was extended from a 6-bit code to an 8-bit code. The added 2 bits are used as additional zone bits, expanding the zone to 4 bits. The resulting code is called the Extended Binary-Coded Decimal Interchange Code (EBCDIC). In this code it is possible to represent 256 (28) different characters instead of 64 (26). In addition to the various character requirements mentioned above, this also allows a large variety of printable characters and several nonprintable control characters. The control characters are used to control such activities as printer vertical spacing, movement of cursor on the terminal screen, etc. All of the 256 bit combinations have not yet been assigned characters, so the code can still grow as new requirements develop. Because EBCDIC is an 8-bit code, it can be easily divided into two 4-bit groups. Each of these 4-bit groups can be represented by 1 hexadecimal digit (refer to Chapter 3). Thus, hexadecimal number system is used as shortcut notation for memory dump by computers that use EBCDIC for internal representation of characters. This results in a one-to-four reduction in the volume of memory dump. Figure 4.3 shows the alphabetic and numerical characters in EBCDIC along with their hexadecimal equivalent. Developed by IBM, EBCDIC code is used in most IBM models and in many other computers. Character EBCDIC Code Hexadecimal Equivalent A B C D E F
Zone 1100 1100 1100 1100 1100 1100
Digit 0001 0010 0011 0100 0101 0110
Cl C2 C3 C4 C5 C6
G H I
1100 1100 1100
0111 1000 1001
C7 C8 C9
J K L M N 0 P Q R
1101 1101 1101 1101 1101 1101 1101 1101 1101
0001 0010 0011 0100 0101 0110 0111 1000 1001
Dl D2 D3 D4 D5 D6 D7 D8 D9
S T U V w X Y z
1110 1110 1110 1110 1110 1110 1110 1110
0010 0011 0100 0101 0110 0111 1000 1001
E2 E3 E4 E5 E6 E7 E8 E9
.
0 mi 0000 F0 1 mi 0001 Fl 2 mi 0010 F2 3 mi 0011 F3 4 nil 0100 F4 5 mi 0101 F5 6 mi 0110 F6 7 nil 0111 F7 8 nil 1000 F8 9 mi 1001 F9 Figure 4.3. Alphabetic and numeric characters in EBCDIC along with their Hexadecimal eauivalent. Zoned and Packed Decimal Numbers From Figure 4.3 it can be observed that in the EBCDIC code, the digit values are the same as the numeric characters - 0 through 9 (0000 - 1001). But numeric values need some special consideration because we must have a way of indicating whether the number is positive, negative, or unsigned (implies positive). Thus, when a numeric value is represented in EBCDIC, a sign indicator is used in the zone position of the rightmost digit. A sign indicator of hexadecimal C is a plus sign, hexadecimal D is a minus sign, and hexadecimal F means the number is unsigned. Figure 4.4 illustrates the
representation of numeric values in EBCDIC. Note that the only zone affected by the sign is the* zone of the rightmost digit. All other zones remain as F, the zone value for numeric characters in EBCDIC. Because each decimal digit has a zone with it, numbers coded in EBCDIC are called zoned decimal numbers. Numeric data input into the computer are usually zoned decimal numbers. Printers can print only those numeric characters that are in a zoned-decimal format. Numeric Value
EBCDIC
Sign Indicator
345 +345 -345
F3F4F5 F3F4C5 F3F4D5
F for unsigned C for positive D for negative
Figure 4.4. Numeric values in EBCDIC in Hexadecimal notation. However, most computers cannot perform arithmetic operations on zoned-decimal data. Thus, before any arithmetic operation can be performed, the data must be converted to a format on which arithmetic operations are possible. One such acceptable format is the packed decimal format. The following steps are used to convert a zoned decimal nunoer to a packed decimal number: Step 1: The zone half and the digit half of the rightmost byte are reversed. This moves the sign to the extreme right of the number. Step 2: All remaining zones are dropped out. Figure 4.5 illustrates the conversion process of zoned decimal data to packed data. Numeric Value Zoned Format Packed Format 345 F3F4F5 345F +345 F3F4C5 345C -345 F3F4D5 345D 3456 F3F4F5F6 03456F Figure 4.5. Zoned and packed decimal numbers. It may be observed that packed data requires fewer number of bytes (group of 8 bits) as compared to zoned data. In the zoned format, there is only one digit per byte (each digit along with the zone requires 8 bits). But there are two digits in each byte in the packed format (each digit requires 4 bits). If the packing process does not completely fill a byte, it is filled with a zero. For example, the zoned data F3F4F5F6 will convert to packed data 03456F. Observe that in this example, the zoned data requires 4 bytes and the packed data requires only 3 bytes Example 4.3. Using binary notation, write the EBCDIC coding for the word BIT. How many bytes are required for this representation? Solution: B = 1100 0010 in EBCDIC binary notation
I = 1100 1001 in EBCDIC binary notation T = 1110 0011 in EBCDIC-binary notation So the EBCDIC coding for the word BIT in binary notation will be 11000010/B 11001001/I 11100011/T 3 bytes will be required for this representation because each letter requires 1 byte (or 8 bits). Example 4.4. Write the EBCDIC coding for the word ZONE (use hexadecimal notation). How many bytes will be required for this representation? Solution: Z = E9 in EBCDIC hexadecimal notation O = D6 in EBCDIC hexadecimal notation N = D5 in EBCDIC hexadecimal notation E = C5 in EBCDIC hexadecimal notation So the EBCDIC coding for the word ZONE in hexadecimal notation will be E9/Z D6/O D5/N C5/E Each hexadecimal digit requires 4 bits and there are altogether 8 hexadecimal digits. So in all 8 x 4 = 32 bits or 4 bytes (8 bits = 1 byte) will be required for this representation. . > We may also write directly that since each letter requires 1 byte for its representation in EBCDIC and there are 4 letters in the word ZONE, 4 bytes will be required for this representation. Example 4.5. Write the EBCDIC zoned-decimal coding for the value +256 (use hexadecimal). How many bytes will be required for this representation? Solution: +256 = F2F5C6 in EBCDIC Each hexadecimal digit requires 4 bits and there are altogether 6 hexadecimal digits. So in all 6 x 4 = 24 bits or 3 bytes (8 bits = 1 byte) will be required for this representation. We may also write directly that since each digit requires 1 byte for its representation in the EBCDIC zoned decimal coding and there are 3 digits in the given number, 3 bytes will be required for this representation. Example 4.6. Write -128 as packed decimal number (use hexadecimal). required for this representation? Solution:
How many bytes will be
-128 = F1F2D8 in EBCDIC = 128D in packed format Each hexadecimal digit requires 4 bits and there are altogether 4 hexadecimal digits. So in all 4 x 4 = 16 bits or 2 bytes (1 byte = 8 bits) will be required for this representation. ASCII Another computer code that is very widely used is the American Standard Code for Information Interchange (ASCII). ASCII has been adopted by several American computer manufacturers as their computers' internal code. This code is popular in data communications, is used almost exclusively to represent data internally in microcomputers, and is frequently found in the larger computers produced by some vendors. ASCII is of two types - ASCII-7 and ASCII-8. ASCII-7 is a 7-bit code that allows 128 (27) different characters. The first 3 bits are used as zone bits and the last 4 bits indicate the digit. Microcomputers using 8-bit byte (group of 8 bits for 1 byte) use the 7-bit ASCII by leaving the leftmost first bit of each byte as a zero. Figure 4.6 shows the alphabetic and numeric characters in ASCII-7 notation. ASCII-8 is an extended version of ASCII-7. It is an 8-bit code that allows 256 (2 8) different characters rather than 128. The additional bit is added to the zone bits. Figure 4.7 shows the alphabetic and numeric characters in ASCII-8 notation. Observe that other than the zone-value differences, ASCII-7 and ASCII-8 are identical. ASCII also uses hexadecimal as its four-to-one shortcut notation for memory dump. Figures 4.6 and 4.7 also show the hexadecimal equivalent of the ASCII notations. Example 4.7. Write the binary coding for the word BOY in ASCII-7. How many bytes are required for this representation? Character
ASCII-7Code
0 1 .2 3 4 5 6 7* 8 9
Zone Oil Oil Oil Oil on on on on on on
Hexadecimal Equivalent Digit 0000 0001 0010 0011 0100 0101 0110 0111 1000 1001
30 31 32 33 34 35 36 37 38 39
A B C D E F G H I J K L M N 0
100 100 100 100 100 100 100 foo 100 100 100 100 100 100 100
0001 0010 0011 0100 0101 0110 0111 1000 1001 1010 1011 1100 1101 1110 mi
41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F
P Q R S T U V W X Y Z 0 1 2 3 4 5 6 7 8 9
101 101 101 101 101 101 101 101 101 101 101 0101 0101 0101 0101 0101 0101 0101 0101 0101
0000 0001 - 0010 0011 0100 0101 0110 0111 1000 1001 1010 0000 0001 0010 0011 0100 0101 0110 0111 1000 1001
50 51 52 53 54 55 56 57 58 59 5A 50 51 52 53 54 55 56 57 58 59
A B C D E F G
1010 1010 1010 1010 1010 1010 1010
0001 0010 0011 0100 0101 0110 0111
Al A2 A3 A4 A5 A6 A7
H I J K L M N O
1010 1010 1010 1010 1010 1010 1010 1010
1000 1001 1010 1011 1100 1101 1110 mi
A8 A9 AA AB AC AD AE AF
P 1011 0000 B0 Q 1011 0001 Bl R 1011 0010 B2 S 1011 0011 B3 T 1011 0100 B4 U 1011 0101 B5 V 1011 0110 B6 W 1011 0111 B7 X 1011 1000 B8 Y 1011 1001 B9 Z 1011 1010 BA Figure 4.7. Numeric and alphabetic characters in ASCII-8 notation along with their Hexadecimal equivalent. Solution: B = 1000010 in ASCII-7 binary notation O = 1001111 in ASCII-7 binary notation Y = 1011001 in ASCII-7 binary notation Hence, the binary coding for the word BOY in ASCII-7 will be 1000010/B 1001111/O 1011001/Y Since each character in ASCII-7 requires one byte for its representation and there are 3 characters in the word BOY, 3 bytes will be required for this representation. Example 4.8. Write the hexadecimal coding for the word GIRL in ASCII-7. required for this representation? Solution: G = 47 in ASCII-7 hexadecimal notation I = 49 in ASCII-7 hexadecimal notation R = 52 in ASCII-7 hexadecimal notation L = 4C in ASCII-7 hexadecimal notation
How many bytes are
Hence, the hexadecimal coding for the word GIRL in ASCII-7 will be 47/G 49/I 52/R 4C/L = GIRL Since each character in ASCII-7 requires one byte for its representation and there are 4 characters in the word GIRL, 4 bytes will be required for this representation.
Example 4.9. Write the binary coding for the word SKY in ASCII-8. How many bytes are required for this representation? Solution: S = 10110011 in ASCII-8 binary notation K = 10101011 in ASCII-8 binary notation Y = 10111001 in ASCII-8 binary notation Hence, the binary coding for the word SKY in ASCII-8 will be 10110011/S 10101011/K 10111001/Y = SKY Since each character in ASCII-8 requires one byte for its representation and there are 3 characters in the word SKY, 3 bytes will be required for this representation. Example 4.10. Write the hexadecimal coding for The word STAR in ASCII-8. How many bytes are required for this representation? Solution: S = B3 in ASCII-8 hexadecimal notation T = B4 in ASCII-8 hexadecimal notation A = Al in ASCII-8 hexadecimal notation R = B2 in ASCII-8 hexadecimal notation Hence, the hexadecimal coding for the word STAR in ASCII-8 will be B3/S B4/T A1/A B2/R = STAR Since each character in ASCII-8 requires one byte for its representation and there are 4 characters in the word STAR, 4 bytes will be required for this representation. COLLATING SEQUENCE The value of an alphanumeric or alphabetic data element is usually the name of some object. Obviously one would not like to perform any arithmetic on such data but one may like to compare them in order to arrange them in some desired sequence. Now, if we compare the alphabetic values A and B, which one will be treated as greater by the computer? For an answer to such questions, it is necessary to have some assigned ordering among the characters used by the computer. This ordering is known as the collating sequence. Collating sequence may vary from one computer system to another depending on the type of computer code used by a particular computer. To illustrate this, let us consider the computer codes already discussed in this chapter. Observe from Figures 4.2 and 4.3 that the zone values of the characters A through 9 decrease in BCD code from the equivalent of decimal 3 down to 0, while the zone values of the characters A through 9 increases in EBCDIC from the equivalent of decimal 12 to 15. This means that a computer, which uses BCD code for its internal representation of characters will treat alphabetic characters (A, B, ..., Z) to be greater than numeric characters (0, 1, ..., 9). On the other hand, a computer, which uses EBCDIC for its internal representation of characters will treat
numeric characters to be greater than alphabetic characters. Similarly, observe from Figure 4.6 and 4.7 that a computer, which uses ASCII for its internal representation of characters will place numbers ahead of letters during a sort (ascending) because the number characters have a zone value that is less than the zone value for letters. However, whatever may be the type of computer code used, in most (not all - in BCD 0 > 9) collating sequences, the following rules are observed: 1. Letters are considered in alphabetic order (A < B < C < ...
be easily divided into two 4-bit groups. Each of these 4-bit groups can be represented by 1 hexadecimal digit. Thus, hexadecimal number system is used as shortcut notation for memory dump by computers that use EBCDIC for internal representation of characters. 5. ASCII (American Standard Code for Information Interchange) is one of the most popular computer codes. It is of two types - ASCII-7 and ASCII-8. ASCII-7 is a 7-bit code, which can represent 128 different characters and ASCII-8 is an 8-bit code, which can represent 256 different characters. Computers, which use ASCII-7 and group of 8 bits for a byte leave the leftmost first bit of each byte as a zero. ASCII also uses hexadecimal as its four-to-one shortcut notation for memory dump. Questions 1. Define the term 'byte'. What is the difference between a bit and a byte? 2. Write the 4-bit BCD code for the following numbers: (a) 25,o (c) 128,0 (b) 64,o (d) 1024,0 3. Using binary notation, show the BCD coding for the following words: (a) BIT (c) CODE (b) BYTE (d) ZERO 4. Using the octal notation, show the BCD coding for the following words: (a) COMPUTER (c) VIDEO (b) INPUT (d) OUTPUT 5. Why was BCD code extended to EBCDIC? 6. How many different characters are possible for the following codes: (a) BCD (c) ASCII-7 (b) EBCDIC (d) ASCII-8 7. Suppose a new computer code is designed that uses 9 bits. How many different characters are possible in this code? 8. Why are octal and hexadecimal shortcut notations used? Identify the shortcut notations used for each of these computer codes: (a) BCD (c) ASCII-7 (b) EBCDIC (d) ASCII-8 9. Why do we have a packed decimal format? How does it differ from a zoned decimal format? 10. Using binary notation, write the EBCDIC coding for the following words: (a) SUN (c) CAT (b) MOON (d) DOG How many bytes are required for each of these representations? 11. Using hexadecimal notation, write the EBCDIC coding for the following words: (a) PROGRAM (c) BYTE (b) OUTPUT (d) OCTAL How many bytes are required for each of these representations? 12. Using hexadecimal notation, write the zoned-decimal coding for the following numbers: (a) 1256 (c) -63 (b) +439 (d) -786 How many bytes are required for each of these representations?
13. Using hexadecimal notation, write the packed-decimal coding for the following numbers: (a) 12915 (c) 872 (b) +9876 (d) -256 How many bytes are required for each of these representations? 14. List out the similarities and differences between 7-bit and 8-bit ASCII. 15. Using binary notation, write the ASCII-7 and ASCII-8 codes for the following words: (a) DRY (c) DAMP (b) WET (d) TERM How many bytes are required for each of these representations? 16. Using hexadecimal notation, write the ASCII-7 and ASCII-8 codes for the following words: (a) PRINT (c) RUB (b) TYPE (d) GIVE How many bytes are required for each of these representations? 17. Explain the meaning of the term "collating sequence". 18. A computer uses EBCDIC as its internal representation of characters. In which order will this computer sort the following strings? (a) ABC (c) 245 (b) 123 (d) ADD 19. A computer uses ASCII. In which order will this computer sort the following strings? (a) BED (d) 128 (b) 512 (e) BAD . (c) ADD 20. Give the full form of the following abbreviations: (a) BCD (b) EBCDIC (c) ASCII
Chapter 5 Computer Arithmetic In Chapter 4 you have seen that computers store numbers, letters, and other characters in coded form that is related to the binary number system. In this chapter you will learn why computers use binary numbers instead of decimal numbers and how the basic arithmetic operations are performed inside the computer using binary numbers. WHY BINARY? You might have observed in Chapter 3 that the use of a smaller base often requires more positions to represent a given value (recall the reason for using octal and hexadecimal notations). For example, 9io = 10012. Here four positions are required instead of one to represent the decimal number 9 in binary form. In spite of this fact, almost all computers use binary numbers. Hence the obvious question that arises is 'Why do we go for binary numbers instead of decimal numbers?' The reasons are as follows: 1. The first reason is that electronic and electrical components, by their very nature, operate in a binary mode. Information is handled in the computer by electronic/electrical components such as transistors, semiconductors, wires, etc. all of which can only indicate two states or conditions - on(l) or 6ff(0). Transistors are either conducting( 1) or nonconducting(O); magnetic materials are either magnetized( 1) or nonmagnetized(O) in one direction or in the opposite direction; a pulse or voltage is present(l) or absent(O) in wire. All information is represented within the computer by the presence or absence of these various signals. The binary number system, which has only two digits (0 and 1), is most suitable for expressing the two possible states. The concept of binary components is illustrated in Figure 5.1. 2. The second reason is that computer circuits only have to handle two binary digits rather than ten decimal digits. This greatly simplifies the internal circuit design of computers, resulting in less expensive and more reliable circuits. 3. Finally, the binary system is used because everything that can be done in decimal number system can also be done in binary number system. How this is achieved has been discussed below. Figure 5.1. Examples of devices that work in binary mode. These devices can only represent two states - on or off, which can represent 1 (yes) or 0 (no). BINARY ARITHMETIC In this section you will see how the four basic arithmetic operations are performed inside a computer using binary numbers. Actually, binary arithmetic is much simpler to learn because binary number system deals with only two digits - 0 and 1. So all binary numbers are made up of only O's and l's and when arithmetic operations are performed on these numbers, the results are also in O's and 1 's only.
Addition Binary addition is performed in the same manner as decimal addition. However, since binary number system has only two digits, the addition table for binary arithmetic is very simple consisting of only four entries. The complete table for binary addition is as follows: 0+0=0 0+1=1 1+0=1 1 + 1=0 plus a carry of 1 to next higher column Carry-overs are performed in the same manner as in decimal arithmetic. Since 1 is the largest digit in the binary number system, any sum greater than 1 requires that a digit be carried over. For instance, 10 plus 10 binary requires the addition of two 1 's in the second position. Since 1 + 1=0 plus a carry-over of 1, the sum of 10 + 10 is 100 in binary. By repeated use of the above rules, any two binary numbers can be added together by adding two bits at a time. The exact procedure is illustrated with the examples given below. Example 5.1. Add the binary numbers 101 and 10 in both decimal and binary form. Solution: Binary - Decimal 101 5 +10 +2 111 7 Add the binary numbers 10011and 1001 in both decimal and binary form. Solution: Binary Carry 11 10011 +1001 11100
Decimal Carry 1 19 +9 28
In this example of binary addition, a carry is generated for first and second columns. Example 5.3.
Add the binary numbers 100111 and 11011 in both decimal and binary form. Solution: Binary Decimal Carry 11111 carry 1 100111 39 +11011 +27 1000010 66 In this example, we face a new situation (1 + 1 + 1) brought about by the carry-over of 1 in the second column. This can also be handled using the same four rules for binary addition. The addition of three 1 's can be broken up into two steps. First we add only two 1 's giving 10(1 + 1 = 10). The third 1 is now added to this result to obtain 11 (a 1 sum with a 1 carry). So we conclude that 1 + 1 + 1 = 1 plus a carry of 1 to next higher column. Subtraction The principles of decimal subtraction can as well be applied to subtraction of numbers in other bases. It consists of two steps, which are repeated for each column of the numbers. The first step is to determine if it is necessary to borrow. If the subtrahend (the lower digit) is larger than the minuend (the upper digit), it is necessary to borrow from the column to the left. It is important to note here that the value borrowed depends upon the base of the number and is always the decimal equivalent of the base. Thus, in decimal, 10 is borrowed; in binary, 2 is borrowed; in octal, 8 is borrowed; in hexadecimal, 16 is borrowed. The second step is simply to subtract the lower value from the upper value. The complete table for binary subtraction is as follows:64 0-0=0 1-0=1 1-1=0 0 - 1 = 1 with a borrow from the next column
.
Observe that the only case in which it is necessary to borrow is when 1 is subtracted from 0. The exact procedure is illustrated with the examples given below. Example 5.4. Subtract 011102froml01012 Solution: Borrow
12 0202 10101 -01110 00111
In the first column, 0 is subtracted from 1. No borrow is required in this case and the result is 1. In the second column, we have to subtract 1 from 0. As seen in the table for
binary subtraction, a borrow is necessary to perform this subtraction. Hence, a 1 is borrowed from the third column, which becomes 2 in the second column because the base is 2. A 1 in the 4s column is equal to 2 in the 2s column. Now, in the second column, we subtract 1 from 2 giving a result of 1. The borrow performed in the second column reduces the 1 in the third column to 0. Hence, in the third column, once again we have to subtract 1 from 0 for which borrow is required. The fourth column contains a 0 and thus has nothing to borrow. Therefore, we have to borrow from the fifth column. Borrowing 1 from the fifth column gives 2 in the fourth column. A 1 in the 16s column equals 2 in the 8s column. Now the fourth column has something to borrow. When 1 of the 2 in the fourth column is borrowed, it becomes 2 in the third column. Now, in the third column, we subtract 1 from 2 giving a result of 1. The borrow performed in the third column reduces the 1 in the fifth column to 0 and the 2 in the fourth column to 1. Hence, subtraction of the fourth column is now 1 from 1, giving 0 and in the fifth column, subtraction is 0 from 0, giving 0. Thus the final result of subtraction is 001112. The result may be verified by subtracting 14i0 (= 011102) from 21,o (101012) which gives 7,o (= 001112). Example 5.5. Subtract 01110002 from 10111002. Solution: Borrow
2 1011100 -0111000 0100100
The result may be verified by subtracting 5610 (= 01110002) from 9210 (= 10111002), which gives 36,0 (= 01001002). Additive Method of Subtraction The direct method of subtraction using the-borrow concept seems to be easiest when we perform subtraction with paper and pencil. However when subtraction is implemented by means of digital components, this method is found to be less efficient than the additive method of subtraction. It may sound surprising that even subtraction is performed using an additive method. This method of subtraction by an additive approach is known as complementary subtraction. In order to understand complementary subtraction, it is necessary to know what is meant by the complement of a number. For a number, which has n digits in it, a complement is defined as the difference between the number and the base raised to the nth power minus one. The definition is illustrated with the following examples: Example 5.6.
Find the complement of 37 Solution: Since the number has 2 digits and the value of base is 10, (Base)"-l = 102-1=99 Now 99 - 37 = 62 Hence, the complement of 3710 = 6210. Example 5.7. Find the complement of 6g. Solution: Since the number has 1 digit and the value of base is 8, (Base)n - 1 = 81 - 1 = 710 Also 68 = 610 Now 710 - 610 = 110 =18 lg Hence, the complement of 68 = l8. Example 5.8. Find the complement of IOIOI2. Solution: Since the number has 5 digits and the value of base is 2, (Base)n-l=25-l=3110 Alsol01012 = 2110 Now 31io- 21io = lO10 = 10102 Hence, the complement of 101012= 010102. Observe from Example 5.8 that in case of binary numbers, it is not necessary to go through the usual process of obtaining complement. Instead, when dealing with binary numbers, a quick way to obtain a number's complement is to transform all its 0's to l's, and all its l's to 0's. For example, the complement of 1011010 is 0100101. Hence, the circuit for obtaining complement of a number in binary system can be easily designed at very less expense. After having seen how to obtain the complement of a number, we will now see how subtraction is performed using the complementary method. Subtraction by the complementary-method involves the following three steps: Step 1: Find the complement of the number you are subtracting (subtrahend); Step 2: Add this to the number from which you are taking away (minuend); Step 3: If there is a carry of 1, add it to obtain the result; if there is no carry, recomplement the sum and attach a negative sign to obtain the result. To illustrate the procedure, let us first consider few examples for decimal numbers.
Example 5.9. Subtract 5610 from 9210 using complementary method. Solution: Step 1: Complement of 5610 = 102-1-56 = 99-56 = 4310 Step 2: 92 +43 (complement of 56) 135 Step 3: 1 (add the carry of 1) Result = 36 The result may be verified using the method of normal subtraction: 92-56 = 36. Example 5.10. Subtract 3510 from 1810 using complementary method. Solution: Step 1: Complement of 3 510 = 102 -1 -35 = 99-35 = 6410 Step 2: 18 +64 (complement of 35) 82 Step 3: Since there is no carry, recomplement the sum and attach a negative sign to obtain the result. Result =-(99-82) = -17 The result may be verified using the method of normal subtraction: 18 - 35 = -17 Let us re-work these examples using binary numbers. Example 5.11. Subtract 01110002 (5610) from 10111002 (9210) using complementary method. Solution: 1011100 +1000111 (complement of 0111000)
10100011 1 (add the carry of 1) 0100100 Result = 01001002 = 3610 Verify the result with the results of Example 5.5 and Example 5.9. Example 5.12. Subtract 1000112(3510) from 0100102 (1810) using complementary method. Solution: 010010 +011100 (complement of 100011) 101110 Since there is no carry, we have to complement the sum and attach a negative sign to it. Hence, Result = -0100012 (complement of 1011102) = -1710 Multiplication Multiplication in the binary number system also follows the same general rules as decimal number system multiplication. However, learning the binary multiplication is a trivial task because the table for binary multiplication is very short, with only four entries instead of the 100 necessary for decimal multiplication. The complete table for binary multiplication is as follows: 0x0=0 0x1=0 1x0=0 1x1=1 The method of binary multiplication is illustrated with the example given below. It is only necessary to copy the multiplicand if the digit in the multiplier is 1, and to copy all 0s if the digit in the multiplier is a 0. The ease with which each step of the operation is .performed is apparent. Example 5.14. Multiply the binary numbers 1010 and 1001. Solution: 1010 Multiplicand x1001 Multiplier
1010 Partial Product 0000 Partial Product 0000 Partial Product 1010 Partial Product 1011010 Final Product Note that the multiplicand is simply copied when multiplier digit is 1 and when the multiplier digit is 0, the partial product is only a string of zero's. As in decimal multiplication, each partial product is shifted one place to the left from the previous partial product. Finally, all the partial products obtained in this manner are added according to the binary addition rules to obtain the final product. In actual practice, whenever a 0 appears in the multiplier, a separate partial product consisting of a string of zeros need not be generated. Instead, only a left shift will do. As a result, Example 5.14 may be reduced to 1010 x l00l 1010 1010SS (S = left shift) 1011010 A computer would also follow this procedure in performing multiplication. The result of this multiplication may be verified by multiplying 1010 (10102) by 910 (10012), which produces a result of 9010 (10110102). It may not be obvious how to handle the addition if the result of the multiplication gives columns with more than two Is. They can be handled as pairs or by adjusting the column to which the carry is placed, as shown by Example 5.15. Example 5.15. Multiply the binary numbers 1111 and 111. Solution:
1 1
1
1 X
1 1
1 1
1 1
1 1 1
1 1
1
1 1
1 1 1
0
1
0
0
1
As shown below, the addition in column 3 above can be handled in any one of the following two ways: Addition handled as pairs (Column 3)
Additive Method of Multiplication Most computers perform multiplication operation by the way of addition only. This can be easily seen by considering an example, say 4x8. The result for this multiplication can be determined by evaluating, with necessary carry overs, 8 + 8 + 8 + 8. That is, the result is obtained simply by adding the digit 8 four times. Similarly, the computer performs all multiplication operations in binary using the additive approach. The idea of repeated addition may seem to be a longer way of doing things, but remember that the computer is well suited to carry out the operations at great speed. The internal circuit design of computer systems is also simplified to a great extent by using this method of multiplication. Division Once again, division in binary number system is very simple. As in the decimal number system (or in any other number system), division by zero is meaningless. Hence, the complete table for binary division is as follows: 0÷1=0 1÷1=1 The division process is performed in a manner similar to decimal division. The rules for binary division are: 1. Start from the left of the dividend. 2. Perform a series of subtractions in which the divisor is subtracted from the dividend. 3. If subtraction is possible, put a 1 in the quotient and subtract the divisor from the corresponding digits of dividend. 4. If subtraction is not possible (divisor greater than remainder), record a 0 in the quotient. 5. Bring down the next digit to add to the remainder digits. Proceed as before in a manner similar to long division. The method is illustrated by Example 5.16. The result may be verified by dividing 3310 (1000012) by 610 (1102), which gives a quotient of 510 (1012) and a remainder of 310 (112). Additive Method of Division Even division operation is performed inside most computers by the process of addition only. This may again sound surprising, but it is true. The computer performs the division operation essentially by repeating the complementary subtraction method. For example, 35 -f 5 may be thought of as: 35-5 = 30 30-5 = 25
25-5 = 20 20-5 = 15 15-5 = 10 10-5 = 5 5-5 = 0 That is, the divisor is subtracted repeatedly from the dividend until the result of subtraction becomes less than or equal to zero. The total number of times subtraction was performed gives the value of the quotient. In this case, the value of quotient is 7 because the divisor (5) has been subtracted 7 times from the dividend (35) until the result of subtraction becomes zero. If the result of last subtraction is zero, then there is no remainder for the division. But, if it is less than zero, then the last subtraction is ignored and the result of the previous subtraction is taken as the value of the remainder. In this case, the last subtraction operation is not counted for evaluating the value of the quotient. The process is illustrated below with an example. Example 5.17. Divide 3310 by 610 using the method of addition. Solution: 33-6 = 27 27-6 = 21 21-6=15 15-6= 9 9-6= 3 3 - 6 = -3 • Total number of subtractions = 6. Since the result of the last subtraction is less than zero, Quotient = 6-1 (ignore last subtraction) = 5 Remainder = 3 (result of previous subtraction) Hence, 33 ÷ 6 = 5 with a remainder 3. Note that it has been assumed here that all the subtraction operations are carried out using the complementary subtraction method (additive method). Once again, performing division inside a computer by the way of addition is desirable because the addition and complementation operations are easily performed in a computer and usually save the labour and expense of designing complicated circuits. We have demonstrated how computer arithmetic is based on addition. Exactly how this simplifies matter can only be understood in the context of binary (not in decimal). The number of individual steps may indeed be increased because all computer arithmetic is reduced to addition, but the computer can carry out binary additions at such great speed that this is not a disadvantage. Points to Remember 1. Almost all computers use binary numbers for internal computations because electronic circuits for performing arithmetic operations in binary mode can be designed
and implemented more easily and reliably at a much lesser cost than those required for performing arithmetic operations in decimal mode. 2. The rules for binary addition are as follows: 0+0=0 0+1=1 1+0=1 1 + 1 = 0 plus a carry of 1 to next higher column 3. The rules for binary subtraction are as follows: 0-0=0 1-0=1 1-1=0 0 - 1 = 1 with a borrow from the next column 4. For a number, which has n digits in it, a complement is defined as the difference between the number and the base raised to the nth power minus one. 5. A quick way to obtain the complement of a binary number is to transform all its 0s to Is, and all its Is to 0s. 6. Complementary subtraction is an additive approach of subtraction. 7. The rules for binary multiplication are as follows: 0x0=0 0x1=0 1x0=0 1x1=1 8. The rules for binary division are as follows: 0÷1=0 1÷1=1 9. Most computers use the additive approach for performing multiplication and division operations. Questions 1. 2. 3. 4. 5. (a) (b) (c) 6. (a) (b) (c) 7. 8. 9.
Why have computers been designed to use the binary number system? Add the binary numbers 1011 and 101 in both decimal and binary form. Add the binary numbers 1010110 and 1011010. Add the binary numbers 10111 and 1011. Find the complement of the following numbers: 49510 (d) C16 2910 (e) 25 48 (f) 324 Find the complement of the following binary numbers: 10 (d) 011011 101 (e) 10110001 101101 (f) 001101001110 Subtract 01101112 from 11011102. Subtract 010102 from 100002. Subtract 0110112 from 1101112.
10. Subtract 2510 from 5010 using complementary method. 11. Subtract 2510 from 2010 using complementary method. 12. Subtract 23410 from 58810 using complementary method. 13. Subtract 21610 from 17210 using complementary method. 14. Subtract 010102 from 100002 using complementary method. 15. Subtract 1101112 from 1011102 using complementary method. 16. Subtract 0110112 from 1101112 using complementary method. 17. Subtract 11112 from 11002 using complementary method. 18. Multiply the binary numbers 1100 and 1010. 19. Multiply the binary numbers 01101 and 1001. 20. Multiply the binary numbers 101111 and 111. 21. Divide 110012 by 1012. 22. Divide 01101112 by 01112. 23. Briefly explain how multiplication and division operations are performed within a computer using additive approach. 24. What is the primary advantage of performing subtraction by the complementary method in digital computers? 25. Discuss the. advantages and disadvantages of performing the various arithmetic operations by additive method in a diaital comouter. Chapter 6 Boolean Algebra and Logic Circuits Boolean algebra that deals with the binary number system, is very useful in designing logic circuits used by the processors of computer systems. In this chapter you will learn about this algebra and the elementary logic gates that are used to build,up circuits of different types to perform the necessary arithmetic operations. These logic gates are the building blocks of all the circuits in a computer. You will also learn how to use Boolean algebra to design simple logic circuits, which are frequently used by the arithmetic logic unit of almost all computers. BOOLEAN ALGEBRA In the mid-1800's, an algebra which simplified the representation and manipulation of propositional logic was developed by the English mathematician, George Boole (18151864). It became known as Boolean algebra after its developer's name. Later, in the year 1938, Claude E. Shannon, a research assistant in the Department of Electrical Engineering at the Massachusetts Institute of Technology, published a thesis entitled, "A Symbolic Analysis of Relay and Switching Circuits". In this thesis, he proposed the use of Boolean algebra in the design of relay switching circuits. The basic techniques described by Shannon were adopted almost universally for the design and analysis of switching circuits. Because of the analogous relationship between the action of relays and of modern electronic circuits, the same techniques, which were developed for the design of relay circuits are still being used in the design of modern computers.
Boolean algebra provides an economical and straightforward approach to the design of relay and other types of switching circuits. Just as an ordinary algebraic expression may be simplified by means of the basic theorems, the expression describing a given switching circuit network may also be reduced or simplified using Boolean algebra. Boolean algebra is now being used extensively in designing the circuitry used in computers. Fundamental Concepts of Boolean Algebra Boolean algebra is based on the fundamental concepts described below. Use of Binary Digits In a normal algebraic expression, a variable can take any numerical value. For example, in the expression 3A + 7B = C, we assume that A, B, and C may range through the entire field of real numbers. Since Boolean algebra deals with the binary number system, the variables used in the Boolean equations may have only two possible values (0 and 1). If an equation describing logical circuitry has several variables, it is still understood that each of the variables can assume only the values 0 or 1. For example, in the equation A + B = C, each of the variables A, B and C may have only the values 0 or 1. Logical Addition The symbol '+' is used for logical addition operator. It is also known as 'OR' operator. We can define the + symbol (OR operator) by listing all possible combinations of A and B and the resulting value of C in the equation A + B = C. It may be noted that since the variables A and B can have only two possible values (0 or 1) so only four (22) combinations of inputs, are possible as shown in Figure 6.1. The resulting output values for each of the four input combinations are given in the table. Such a table is known as a truth table. Thus, Figure 6.1 is the truth table for the logical OR operator. Observe that the result is 0 only when the value of both the input variables is 0. The result is 1 when any of the input variables is 1. Note that a result of 1 is also obtained when both the inputs A and B are 1. This is the reason why the + symbol does not have the "normal" meaning, but is a logical addition operator. This concept of logical addition may be extended to any number of variables. For example, in the equation A + B + C + D = E, even if A, B, C, and D all had the value of 1, the sum of the values (the result E) would be 1 only. The equation A + B = C is normally read as "A or B equals C". Inputs A 0 0 1 1
Output +
B 0 1 0 1
C 0 1 1 1
Figure 6.1. Truth table for logical OR (+) operator. Logical Multiplication The symbol '.' is used for logical multiplication operator. It is also known as 'AND' operator. We can again define the • symbol (AND operator) by listing all possible combinations of A and B and the resulting value of C in the equation A • B = C. The truth table for logical AND operator is shown in Figure 6.2. Observe from the truth table that the result C is equal to 1 only when both the input variables A and B are 1, otherwise it is 0. The equation A.B=C is normally read as “ A and B equals C”. Inputs
Output
A • B C 0 0 0 0 1 0 1 0 0 1 1 1 Figure 6.2. Truth table for logical AND (•) operator. Complementation The two operations defined so far (OR and AND) are binary operations because they define an operation on two variables. The complementation operation is a unary operation, which is defined on a single variable. The symbol '-' is normally used for complementation operator. It is also known as 'NOT' operator. Thus we write A, meaning "take the complement of A", or (A + B), meaning "take the complement of A + B." The complementation of a variable is the reverse of its value. Thus, if A = 0 then A = 1 and if A = 1 then A = 0. The truth table for logical NOT (-) operator is shown in Figure 6.3. "complement of A" or "not of A". Tnput
Output
A
A
0 1
1 0
Figure 6.3. Truth table for logical NOT (-) operator. Operator Precedence
A is read as
Does A + B • C mean (A + B) • C or A + (B • C)? The two generate different values for A = 1, B = 0, and C = 0, for then we have (1 + 0) • 0 = 0 and 1 + (0 ■ 0) = 1, which differ. Hence it is necessary to define operator precedence in order to correctly evaluate boolean expressions. The precedence of Boolean operators is as follows: 1. 2. 3. 4. 5.
The expression is scanned from left to right. Expressions enclosed within parentheses are evaluated first. All complement (NOT) operations are performed next. All '•' (AND) operations are performed after that. Finally, all'+' (OR) operations are performed in the end.
So according to this precedence rule, A + B • C means A + (B • C). Similarly for the expression A • B, the complement of A and B are both evaluated first and the results are then ANDed. Again for the expression [A + BJ, the expression inside the parenthesis (A + B) is evaluated first and the result so obtained is then complemented. Postulates of Boolean Algebra Postulate 1: (a) A = 0 if and only if A is not equal to 1 (b) A = 1 if and only if A is not equal to 0 Postulate 2: (a) x + 0 = x (b) x . 1 = x Postulate 3: Commutative Law (a) x + y = y + x (b) x • y = y • x Postulate 4: Associative Law (a) x + (y + z) = (x + y) + z (b) x . (y • z) = (x • y) • z Postulate 5: Distributive Law (a) x • (y + z) = x • y + x • z (b) x + y . z = (x + y) • (x + z) Postulate 6: (a) x + x = 1 (b) x • x = 0 The postulates listed above are the basic axioms of the algebraic structure and need no proof. They are used to prove the theorems of Boolean algebra. The Principle of Duality In Boolean algebra, there is a precise duality between the operators • (AND) and + (OR), and the digits 0 and 1. For instance, let us consider Figure 6.4. We can see that the second row of the table is obtainable from the first row and vice-versa simply by interchanging '+' with '•' and '0' with ' 1'. This important property is known as the principle of duality in Boolean algebra.
Column 1
Column Column 3 2 Row 1 1 + 1 = 1 1 + 0 = 0+ = 1 0 + 0 = 0 = 1= Row 2 00 = 0 0- 1 = 1 0 1-1 = 1 = -0 = Figure 6.4. Illustrating the principle of duality in Boolean algebra. The implication of this principle is that any theorem in Boolean algebra has its dual obtainable by interchanging '+' with '•' and '0' with ' 1'. Thus if a particular theorem is proved, its dual theorem automatically holds and need not be proved separately. Theorems of Boolean Algebra Some of the important theorems of Boolean algebra are stated below along with their proof. Theorem 1 (Idempotent law) (a) x + x = x (b) x • x = x Proofof(a)
Proof of (b)
L.H.S.
L.H.S.
=X+X
=x•x
= (x + x) • 1
by postulate 2(b)
=x-x+0
by postulate 2(a)
by postulate 6(a)
=X■X+X• X
by postulate 6(b)
=X+X• X
by postulate 5(b)
= x • (x + x )
by postulate 5(a)
=x+0
by postulate 6(b)
=X•1
by postulate 6(a)
=x
by postulate 2(a)
=x
by postulate 2(b)
= (x + x) • (x +
= R.H.S.
X)
= R.H.S.
Note that Theorem l(b). is the dual of Theorem l(a) and that each step of the proof in part (b) is the dual of part (a). Any dual theorem can be similarly derived from the proof of its corresponding pair. Hence from now onwards, the proof of part (a) only will be given. Interested readers can apply the principle of duality to the various steps of the proof of part (a) to obtain the proof of part (b) for any theorem. Theorem 3 (Absorption law) (a) x + x . y = x (b) x • (x + y) = x Proof of (a)
Proof by the Method of Perfect Induction The theorems of Boolean algebra can also be proved by means of truth tables. In a truth table, both sides of the relation are checked to yield identical results for all possible combinations of variables involved. In principle, it is possible to enumerate all possible combinations of the variables involved because Boolean algebra deals with variables that can have only two values. This method of proving theorems is called exhaustive enumeration or perfect induction. For example, Figure 6.5 is a truth table for proving Theorem 3(a) by the method of perfect induction. Similarly, Figure 6.6 proves Theorem 3(b) by the method of perfect induction. X y x-y x+x■y 0 0 1 1
0 l 0 I
0 0 0 1
0 0 1 1
Figure 6.5. Truth table for proving Theorem 3(a) by the method of perfect-induction. X 0 0 1 1
y
x+y
x ■ (x + y)
0 l 0 l
0 1 1 1
0 0 1 1
Figure 6.6. Truth table for proving Theorem 3(b) by the method of perfect induction. Theorem 4 (Involution Law) x=x Proof Figure 6.7 proves this theorem by the method of perfect induction. X
X
X
0 1
1
0 i
Note that Theorem 4 has no dual since it deals with the NOT operator which is unary operator.
Theorem 5 (a) x-(x +y) = x-y (b) x + x-y = x + y Proof of (a) Figure 6.8 proves this theorem by the method of perfect induction. X y X x+y x-(x + y) x-y 0 0 1 1
0 l 0 1
1 1 0 0
1 1 0 1
0 0 0 1
_j
0 0 0 1
Figure 6.8. Truth table for proving Theorem 5(a) by the method of perfect induction. Proof of (b) Figure 6.9 proves this theorem by the method of perfect induction. X y X X- y x+ \ • y x+y 0 0 1 1
0 l 0 l
1 1 0 0
0 1 0 0
0 1 1 1
0 1 1 1
Figure 6.9. Truth table for proving Theorem 5(b) by the method of perfect induction. Theorem 6 (De Morgan's Law) (a) x+y = x • y (b) x • y = x + y Proof of (a) ' Figure 6.10 proves this theorem by the method of perfect induction. X y x+y X X x+y y y 0 0 1 1
0 l 0 l
0 1 1 1
1 0 0 0
1 1 0 0
l 0 l 0
1 0 0 0
Figure 6.10. Truth table for proving Theorem 6(a) by the method of perfect induction. Proof of (b) Figure 6.11 proves this theorem by the method of perfect induction.
X
y
x • y-
0 0 1 1
0
0 0 0 1
I
0 I
x-y '-y 1 1 1 0
X'
y
x+y
1 1 0 0
l 0 l 0
1 1 1 0
Figure 6.11. Truth table for proving Theorem 6(b) by the method of perfect induction. Theorems 6(a) and 6(b) are important and very useful. They are known as De Morgan's laws. They can be extended to n variables as given below: X1+X2+X3+... + Xn = X1-X2-X3-...-X11 x,.x2-x3.... xn=x1+x2 The basic Boolean identities are summarized in Figure 6.12. It is suggested that the readers should become well conversant with the identities given in this table in order to use the algebra effectively. Sr. No.
Identities
Dual identities
1
A+0=A
A- 1 =A
2
A+l = 1
A-0 = 0
3
A«+A = A
A-A = A
4
A + A =1
A- A =0
5
A=A
6
A+B=B+A
7
9
(A + B) + C = A + (B + (A • B) • C = A • (B • C) C) A(B + C) = AB + A-C A + B ■ C = (A + B) • (A + C) A+A-B=A A- (A + B) = A
10
A+ A -B = A + B
A-(A+B) = A-B
11
A + B= A•B
AB= A + B
8
AB=B-A
Figure 6.12. Summary of basic Boolean identities.
BOOLEAN FUNCTIONS
A Boolean function is an expression formed with binary variables, the two binary operators OR and AND, the unary operator NOT, parentheses and equal sign. For a given value of the variables, the value of the function can be either 0 or 1. For example, consider the equation W = X+ Y Z Here the variable W is a function of X, Y and Z. This is written as W = f(X, Y, Z) and the right hand side of the equation is called an expression. The symbols X, Y and Z are referred to as literals of this function. The above is an example of a Boolean function represented as an algebraic expression. A Boolean function may also be represented in the form of a truth table. The number of rows in the table will be equal to 2n, where n is the number of literals (binary variables) used in the function. The combinations of 0s and Is for each row of this table is easily obtained from the binary numbers by counting from 0 to 2" - 1. For each row of the table, there is a value for the function equal to either 0 or 1, which is listed in a separate column of the table. Such a truth table for the function W = X + Y • Z is shown in Figure 6.13. Observe that there are eight (23) possible distinct combinations for assigning bits to three variables. The column labeled W is either a 0 or a 1 for each of these combinations. The table shows that out of eight, there are five different combinations for which W = 1. X Y Z w u 0 0 0 1 1 1 1
0 0 1 1 -"0 0 1 1
0 1 0 1 0 1 0 1
0 1 0 0 r l l l
Figure 6.13. Truth table for the Boolean function W = X + Y • Z. The question now arises - is an algebraic expression for a given Boolean function unique? In other words, is it possible to find two algebraic expressions that specify the same Boolean function? The answer to this question is yes. As a matter of fact, the manipulation of Boolean algebra is applied mostly to the problem of finding simpler expressions for a given expression. For example, let us consider the following two functions: F1= x-y-z+x-y-z+x-y and F2 = x . y+ x-z The representation of these two functions in the form of truth table is shown in Figure 6.14. From the table we find that the function F2 is the same as the function Fi since both
have identical 0s and Is for each combination of values of the three binary variables x, y and z. In general, two functions of n binary variables are said to be equal if they have the same value for all possible 2n combinations of the n literals. X
y
z
F1
F2
0 0 0 0 1 1 1 1
0 0 l l 0 0 l l
0 1 0 1 0 1 0 1
0 1 0 1 1 1 0 0
0 1 0 1 .1 1 0 0
Figure 6.14. Truth table for the Boolean functions: F1 = x -y z +x -y-z+x-y and F2 = xy + xz Minimization of Boolean Functions When a Boolean function is implemented with logic gates (discussed later in this chapter), each literal in the function designates an input to a gate and each term is implemented with a gate. Thus for a given Boolean function, the minimization of the number of literals and the number of terms will result in a circuit with less components. For example, since functions Fi and F2 of Figure 6.14 are equal Boolean functions, it is more constitutes the best form of Boolean function depends on the particular application. However, we will give consideration only to the criterion of component minimization, which is achieved by literal minimization. There are several methods used for minimizing the number of literals in a Boolean function. However, a discussion of all these methods is beyond the scope of this book. Hence here we will consider only the method of algebraic manipulations. Unfortunately, in this method, there are no specific rules or guidelines to be followed that will guarantee the final answer. The only method available is cut-and-try procedure employing the postulates, the basic theorems, and any other manipulation method, which becomes familiar with use. The following examples illustrate this procedure. Example 6.1. Simplify the following Boolean functions to a minimum number of literals, (a) x + x • y (b)x-(x + y) (c) x • y • z + x • y- z + x • y (d) x • y + x • z + y- z (e) (x + y)-(x+z)-(y + z)
(d)x-y+x-z + y-z = x-y+x-z + y-z-l = x-y+x-z + y-z-(x+x) = xy+xz + yzx + yz- x =x■y+x•z+x■y•z+x•y■z = x-y-l + x-z + x-y-z+x-y-z = x-y-1+x-y-z+x-z+x-y-z = x-y-(l +z)+ x -z-(l +y) = x-y-(z+ 1)+ x -z-(y+ 1) =x•y•1+x•z■1 = x -y+ x ■ z (e) (x + y)-(x +z)-( = (x + y)-(x + z) Note that in Example 6.1, functions (a) and (b) are the dual of each other and use dual expressions in corresponding minimization steps. Function (c) shows the equality of the functions Fi and F2 of Figure 6.14. Function (d) illustrates the fact that an increase in the number of literals sometimes leads to a final simpler expression. Observe that function (e) is the dual of function (d). Hence it is not minimized directly and can be easily derived from the dual of the steps used to derive function (d). Complement of a Function The complement of a Boolean function F is F and is obtained by interchanging O's for 1 's and 1 's for O's in the truth table that defines the function. For example, Figure 6.15 defines the function F = x y + x • z and its complement F. X
y
z
F
F
0 0 0 0 1 0 0 1 1 0 0 1 0 0 1 0 1 1 1 0 1 0 0 1 0 1 0 1 1 0 1 1 0 0 1 ] 1 1 0 1 Figure 6.15. Truth table for the function F = x • y + X • z and its comolemont Algebraically, the complement of a function may be derived through De Morgan's theorems whose generalized forms are as follows: A1+A2+A3+... + An = A1.A2.A3 ….An A1 -A2 • A3 •... • An = A, + A2 + A3 +.. + An
These theorems state that the complement of a function is obtained by interchanging the OR and the AND operators and complementing each literal. The method is illustrated below through an example. Example 6.2. Find the complement of the following functions: (a) F, = x • y • z + x • y • z (b) F2=x-(y-z+yz) Solution: Applying De Morgan's theorems as many times as necessary, the complements are obtained as follows: (a) F, = x • y • z + x • y • z =(x-y-z)-(x-y-z) = (x + y + z) ■ (x + y + z) (b) F2=x-(y-z + y x + (y-z)-(y-z) A simpler procedure for deriving the complement of a function is to take the dual of the function and then complement each literal. This method follows from the generalized De Morgan's theorems. Remember that the dual of a function is obtained by interchanging OR and AND operators and O's and l's. The method is illustrated below with the help of an example. Example 6.3. Find the complement of the functions Fj and F2 of Example 6.2 by taking their dual and complementing each literal. Solution: (a) F, = xyz + xy-z The dual of F1 is: (x + y + z;) ■ (x + y + z) Complementing each literal we get F2 = (x + y + z)-(x (b) F2 = x-(y-z + y Thedualof F2 is: x + (y + z)-(y + z) Complementing each literal we get + z)-(y.(y+z) Canonical Forms for Boolean Functions Minterms and Maxterms A binary variable may appear either in its normal form (x) or in its complement form (x). Now consider two binary variables x and y combined with an AND operator. Since each variable may appear in either form, there are four possible combinations: x-y, x-y, x-y, x-y
Each of these four AND terms is called a minterm or a standard product. In a similar manner, n variables can be combined to form 2n minterms. The 2n different minterms may be determined by a method similar to the one shown in Figure 6.16 for three variables. The binary numbers from 0 to 2n - 1 are listed under the n variables. Each minterm is obtained from an AND term of the n variables, with each variable being primed if the corresponding bit of the binary number is 0 and unprimed if it is a 1. Variables
Minterms
Maxterms
X
y
z
Term
Designation
Term
Designation
0
0
0
x-y-z
m0
x+y+z
Mo
0 0
0 1
m, m2
x+y+z x+y+z
M, M2
0
1
1 x-y-z 1*1 x-y-z 0 1 x-y-z
m3 -
x+y+z
M3
1 1
0 0
0 1
x-y-z x-y-z
m4 m5
x+y+z x+y+z
M4 M5
1
1
0
x-y-z
m
x+y+z
M6
1
1
1
x-y-z
x+y+z
M7
6 m7
Figure 6.16. Minterms and Maxterms for three variables. A symbol for each minterm is also shown in the figure and is of the form nij, where j denotes the decimal equivalent of the binary number of the minterm designated. In a similar fashion, n variables forming an OR term, with each variable being primed or unprimed, provide 2n possible combinations called maxterms or standard sums. The eight maxterms for three variables, together with their symbolic designation, are listed in Figure 6.16. Any 2n maxterms for n variables may be determined similarly. Each maxterm is obtained from an OR term of the n variables, with each variable being unprimed if the corresponding bit is a 0 and primed if it is a 1. Note that each maxterm is the complement of its corresponding minterm and vice-versa. Sum-of-Products A sum-of-products expression is a product term (minterm) or several product terms (minterms) logically added (ORed) together. For example, the expression x-y + x-y is a sum-of-products expression. The following are all sum-of-products expressions: x
x+y x+y•z x•y+z x•y+x•y•z The following steps are followed to express a Boolean function in its sum-of-products form: 1. Construct a truth table for the given Boolean function. 2. Form a minterm for each combination of the variables which produces a 1 in the function. 3. The desired expression is the sum (OR) of all the minterms obtained in step 2. For example, in case of function Fi of Figure 6.17, the following three combinations of the variables produce a 1: 001, 100, and 111 Their corresponding minterms are x-y-z, x-y-z, and x-y-z Hence, taking the sum (OR) of all these minterms, the function Fi can be expressed in its sum-of-products form as: F1 = x-y-z + x-y-z + x-y-z or F2 = m, + m4 + m7 Similarly, it may be easily verified that the function F2 of Figure 6.17 can be expressed in its sum-of-products form as: F2 = x-y-z + x-y-z + x-y-z+x-y-z X y z 0 0 0 0 1 1 1 1
0 0 l I
0 0 1 1
0 1 0 1 0 1 0 1
or F2 = m3 + m5 + m6 + m7 F1 F2 0 1 0 0 1 0 0 1
0 0 0 1 0 1 1 1
Figure 6.17. Truth table for functions F1and F2. It is sometimes convenient to express a Boolean function in its sum-of-products form. If not in this form, it can be made so by first expanding the expression into a sum of AND terms. Each term is then inspected to see if it contains all the variables. If it misses one or more variables, it is ANDed with an expression of the form (x + x), where x is one of the missing variables. The following example clarifies this procedure. Example 6.4.
Express the Boolean function F = A+ B-C in the sum-of-minterms (products) form. Solution: The function has three variables A, B and C. The first term A is missing two variables, therefore A = A • (B + B) = AB + AB This is still missing one variable, therefore A = A • B • C+C) + A•B• (C+C) =A•B•C•+A•B•C+A•C+A•B•C The second term B • C is missing one variable, therefore B • C = B • C • (A+A) = A • B • C+ A • B • C Hence by combining all the terms we get F = A•B•C + A•B•C +A•B • C + A•B • C +A•B • C+A •B • C But in the above expression, the term ABC appears twice and according to theorem l(a) we have x + x = x. Hence it is possible to remove one of them. Rearranging the minterms in ascending order, we finally obtain: F=A•B•C+A•B•C+A•B•C+A•B•C = m1 m4 + m5 + m6 + m7 It is sometimes convenient to express the Boolean function, when in its sum-of-minterms, in the following short notation: F(A,B,C) = Z(1,4,5,6,7) The summation symbol 'Z' stands for the ORing of terms. The numbers following it are the minterms of the function. Finally, the letters in parentheses with F form a list of the variables in the order taken when the minterm is converted to an AND term. Product-of-Sums A product-of-sums expression is a sum term (maxterm) or several sum terms (maxterms) logically multiplied (ANDed) together. For example, the expression (x +y)-(x+ y) is a product of sums expression. The following are all product-of-sums expressions: x (X+y) (x + y)-z (x+ y)-(x+y)-(x + y) (x + y)-(x+y + z) The following steps are followed to express a Boolean function in its product-of-sums form:
1. Construct a truth table for the given Boolean function. 2. Form a maxterm for each combination of the varibles which produce a 0 in the function. 3. The desired expression is the product (AND) of all the maxterms obtained in step 2. For example in case of function Fi of Figure 6.17, the following five combinations of the variables produce a 0: 000, 010,011, 101, and 110 Their corresponding maxterms are: (x + y + z), (x + y + z), (x + y + z), (x + y + z), and (x + y + z) Hence, taking the product (AND) of all these maxterms, the function Fi can be expressed in its product-of-sums form as: F1 = (x + y + z) • (x + y + z) • (x + y + z) • (x + y + z) • (x + y + z) or F1 =M0-M2-M3-M5-M6 Similarly, it may be easily verified that the function F2 of Figure 6.17 can be expressed in its product-of-sums form as: F2 = (x + y + z)-(x + y + z) • (x + y + z) • (x + y + z) or F2 = M0 • M, • M2 • M4 In order to express a Boolean function in its product-of-sums form, it must first be brought into a form of OR terms. This may be done by using the distributive laws: x + y • z = (x + y) • (x + z) Then any missing variable (say x) in each OR term is ORed with the form x • x. This procedure is explained with the following example. Example 6.5. Express the Boolean function F = x • y + x • z in the product-of-maxterms (sums) form. Solution: At first we convert the function into OR terms using the distributive law: F=x•y+x•z = (x • y + x) • (x • y + z) = (x+x) • (y+x) • (x + z) • (y + z) = (x+y) • (x + z) • (y + z) The function has three variables x, y and z. Each OR term is missing one variable, therefore: x +y = x +y + Z • Z = (x +y + Z) • (X +y+ Z) x + z = x + z +"-y • y =(x + z + y) • (x + z+ y) = (x +y + z)-(x + y +z) y + z = x • +y + z = (x + y + z)-(x+y + z) Combining all the terms and removing those that appear more than once, we finally obtain: F = (x + y + z) • (x + y + z) • (x + y + z) • (x + y+ z)
= M0 • M2 • M4 • M5 A convenient way to express this function is as follows: F (x, y, z) = II (0, 2, 4, 5) The product symbol FI denotes the ANDing of maxterms. The numbers following it are the maxterms of the function. The sum-of-products and the product-of-sums form of Boolean expressions are known as standard forms. One prime reason for liking the sum-of-products or the product-of-sums expressions is their straightforward conversion to very nice gating networks which are more desirable from most implementation points of view. In their purest, nicest form they go into two-level networks, which are networks for which the longest path through which the signal must pass from input to output is two gates. Conversion Between Canonical Forms The complement of a function expressed as the sum-of-minterms equals the sum-ofminterms missing from the original function. This is because the original function is expressed by those minterms that make the function equal to 1, while its complement is a 1 for those minterms for which the function is a 0. For example, the function F (A, B, C) = (1,4, 5,6, 7) = nij +m4 +m5+m6 +m7 has a complement that can be expressed asf F (A, B, C) = (0, 2, 3) = m0 +m2 +m3 Now, if we take the complement of F, by De Morgan's theorem we obtain F back in a different form: F = m0 +m2 +m3 = m0 • m2 • m3 = M0-M2-M3 = II (0, 2, 3) The last conversion follows from the definition of minterms and maxterms as shown in Figure 6.16. From the figure, it is clear that the following relation holds true: That is, the maxterm with subscript/ is a complement of the minterm with the same subscript/, and vice-versa. The last example has demonstrated the conversion between a function expressed in sumof-minterms and its equivalent in product-of-maxterms. A similar argument will show that the conversion between the product-of-maxterms and the sum-of-minterms is similar. We now state a general conversion procedure:
"To convert from one canonical form to another, interchange the symbol and list those numbers missing from the original form." For example, the function F (x, y, z) = TI (0, 2, 4, 5) is expressed in the product-of-maxterms form. Its conversion to sum-of-minterms is: F(x,y,z) = I(l,3,6,7) Note that in order to find the missing terms, one must realize that the total number of minterms or maxterms is always 2n, where n is the number of binary variables in the function. LOGIC GATES All operations within a computer are carried out by means of combinations of signals passing through standard blocks of built-in circuits that are known as logic gates. In other words, a logic gate is an electronic circuit which operates on one or more input signals to produce standard output signals. These logic gates are the building blocks of all the circuits in a computer. Computer circuits are built up using combinations of different types of logic gates to perform the necessary operation. There are several types^of gates, but we shall consider here only some of the most important ones. These are sufficient to introduce the concept of circuit design using logic gates. AND Gate An AND gate is the physical realization of the logical multiplication (AND) operation. That is, it is an electronic circuit that generates an output signal of 1 only if all input signals are also 1. To have a conceptual idea, let us consider the case of Figure 6.18. Here two switches A and B are connected in series. It is obvious that the input current will reach the output point only when both the switches are in the on(l) state. There will be no output (output = 0) if either one or both the switches are in the off(0) state. Hence, two or more switches connected in series behave as an AND gate. Input Output The behaviour of a logic gate, that is the state of its output signal depending on the various combinations of input signals, is conveniently described by means of a truth table. The truth table and the block diagram symbol for an AND gate for two input signals are shown in Figure 6.19. Since there are only two inputs (A and B), only four (22) combinations of inputs are possible. Also observe from the truth table that an output of 1 is obtained only when both the inputs are in 1 state, otherwise it is 0.
Inputs
Output
A
B
C = AB
0 0 1 1
0 1 0 1
0 0 0 1
Figure 6.19. Block diagram symbol and truth table for an AND gate. OR Gate An OR gate is the physical realization of the logical addition (OR) operation. That is, it is an electronic circuit that generates an output signal of 1 if any of the input signals is also 1. Two or more switches connected in parallel behave as an OR gate. It can be seen from Figure 6.20 that the input current will reach the output point when any one of the two switches are in the on(l) state. There will be no output only when both the switches»(A andJB) are in the off(0) state. The truth table and the block diagram symbol for an OR gate for two input signals are shown in Figure 6.21. Observe that an output of 1 is obtained when any of the input signals is 1. Output is 0 only when both the inputs are 0. Inputs Output A 0 0 1 1
8 0 1 0 1
C = A +B 0 1 1 1
Figure 6.21. Block diagram symbol and truth table for an OR gate. Just as the + and ■ operations could be extended to several variables using the associative law, AND gates and OR gates can have more than two inputs. Figure 6.22 shows three input AND and OR gates and the table of all input combinations for each. As might be hoped, the output of the AND gate with inputs A, B and C is a 1 only if A and B and C are 1, i.e., when all three of the inputs are 1, so that we write the output as A • B • C. Similarly, the OR gate with inputs A, B, and C has a 1 output if A or B or C is a 1, so that we can write A + B + C for its output. The above argument can be extended. A four-input AND gate has a 1 output only when all four inputs are 1, and a four-input OR gate has a 1 output when any of its inputs is a 1. ABC
D = ABC Inputs
Output
A B 0 0 0 0 0 1 0 1 1 0 1 0 1 1 1 1 (a) Three input AND gate.
C 0 1 0 1 0 1 0 1
D=A+B+C Inputs A B 0 0 0 0 0 1 0 1 1 0 1 0 1 1 1 1 (b) Three input OR gate.
D = ABC 0 0 0 0 0 0 0 1
Output c 0 1 0 1 0 1 0 1
D=A+B+C 0 1 1 1 1 1 1 1
Figure 6.22. Block diagram symbol and truth table for (a) Three input AND gate (b) Three input OR gate. NOT Gate A NOT gate is the physical realization of the complementation operation. That is, it is an electronic circuit that generates an output signal, which is the reverse of the input signal. A NOT gate is also known as an inverter because it inverts the input. The truth table and the block diagram symbol for a NOT gate are shown in Figure 6.23. Recall that the complementation operation is unary operation, which is defined on a single variable. Hence a NOT gate always has a single input. Figure 6.23 also shows that connecting two NOT gates in series gives an output equal to the input, and this is the gating counterpart to the law of the double complementation, A = A.
Input
Output
A A 0 1 1 0 Figure 6.23. Block diagram symbol and truth table for (a) A NOT gate, and (b) Two NOT gates in series. NAND Gate A NAND gate is a complemented AND gate. That is, the output of NAND gate will be a 1 if any one of the inputs is a 0 and will be a 0 only when all the inputs are 1. The truth table and the block diagram symbol for a NAND gate are shown in Figure 6.24. The symbol ' T ' is usually used to represent a NAND operation in boolean expressions. Thus, A ↑ B= A • B = A + B Inputs
Output
A B C=A+B 0 0 1 0 1 1 1 0 1 1 1 0 Figure 6.24. Block diagram symbol and truth table for a NAND gate. The operation of a NAND gate can be analysed using the equivalent block diagram circuit shown in Figure 6.25, which has an AND gate followd by a NOT gate . For inputs A and B, the output of the AND gate will be A • B, which is fed as input to the NOT gate. Hence the complement of A • B will be A • B, which is equal to A + B or A t B. In fact, the small circle on the output of the NAND gate. Multiple-input NAND gates can be analyzed similarly. A three-input NAND gate with inputs A, B, and C will have an output equal to ABC or A + B + C, which says that the output will be a 1 if any of the inputs is a 0 and will be a 0 only when all three inputs are 1.
NOR Gate
A NOR gate is a complemented OR gate. That is, the output of a NOR gate will be a 1 only when all inputs are 0 and it will be a 0 if any input represents a 1. The truth table and the block diagram symbol for a NOR gate are shown in Figure 6.26. The symbol '↓' is usually used to represent a NOR operation in Boolean expressions. Thus, A ↓ B = A + B = A .B Inputs
Output
A
B
C = A-B
0 0 1 1
0 1 0 1
1 0 0 0
Figure 6.26. Block diagram symbol and the truth table for a NOR gate. The operation of a NOR gate can be analyzed using the equivalent block diagram circuit shown in Figure 6.27, which has an OR gate followed by a NOT gate. For inputs A and B, the output of the OR gate will be A + B, which is fed as input to the NOT gate. Hence the complement of A + B will be A + B, which is equal to A • B or A * B . In fact, the small circle on the output of the NOR gate (see Figure 6.26) represents complementation. The NOR gate can then be seen to be an OR gate followed by a NOT gate. Multiple input NOR gates can be analyzed similarly. A three-input NOR gate with inputs A, B, and C will have an output equal to A + B + C or A-B-C, which says that the output will be a 1 only when all the three inputs are 0 and it will be a 0 if any of the three inputs is a 1. LOGIC CIRCUITS The logic gates described in the previous section are seldom used by themselves but are used in combinations. They are interconnected to form gating/logic networks, which are known as combinational logic circuits. For these logic circuits, the Boolean algebra expression can be derived by systematically progressing from input to output on the gates. Few examples are given below. Find the Boolean expression for the output of the logic circuit given below. Solution: Input A is fed to the.NOT gate whose output will be A. Inputs B and C are fed to the OR gate whose output will be B + C. Now these two outputs (A and B + C) are fed as input to the AND gate. The output produced by the AND gate will be A • (B + C). Hence, D = A • (B + C), which is the required Boolean expression for the output of the given logic circuit.
Example 6.7. Find the logic equation for the output produced by the logic circuit given below. Solution: The output of the OR gate is A + B -------------- (a) The output of the first AND gate is A • B --------------- (b) Since the expression (b) is fed as input to the NOT gate, the output of the NOT gate is A • B -------------- (c) Now expressions (a) and (c) are fed as input to the second AND gate, whose output will be (A + B) • (A•B) Hence, C = (A + B) • (A-B), which is the desired logic equation for the output produced by the given logic circuit. Example 6.8. Find the Boolean expression for the output of the logic circuit given below. Solution: At point 1, the output of the OR gate is A + B __________ (a) At point 2, the output of the NOT gate is C _________ (b) At point 3, the output of the NOT gate is D ________ (c) The inputs to the AND gate at point 4 are (A + B), C, and D. Hence at point 4, the output of the AND gate is (A+B) • C • D ____ (d) The inputs to the AND gate at point 5 are (A + B), C, and D. Hence at point 5, the output of the AND gate is (A + B) •C •D -----------(e) Finally, the inputs to the OR gate at point 6 are (d) and (e). Hence at point 6, the output of the OR gate is (A+B) • C • D+(A + B) • C • D Hence, E = (A + B) • C • D+(A + B) • C • D, which is the required Boolean expression for the output of the given logic circuit.
Converting Expressions to Logic Circuits We have just now considered few examples that illustrate the method of deriving Boolean expression for a given logic circuit. The reverse problem of constructing a logic
circuit for a given Boolean expression is also not difficult. The three logic gates - AND, OR, and NOT are said to be logically complete because any Boolean expression may be realized using only these three gates. The method of constructing logic circuits for Boolean expressions using only these three gates is illustrated below with the help of some examples. The Universal NAND Gate We have seen that AND, OR, and NOT gates are logically complete in the sense that any Boolean function may be realized using these three gates. However, the NAND gate, which was introduced in the previous section, is said to be an universal gate because it is alone sufficient to implement any Boolean function. To show that any Boolean function can be implemented with the sole use of NAND gates, we need to only show that the logical operations AND, OR, and NOT can be implemented with NAND gates. This is shown in Figure 6.28 below. A NOT operation is obtained from a one-input NAND gate. Hence, a single-input NAND gate behaves as an inverter. The AND operation requires two NAND gates. The first one produces the inverted AND and the second one being a single input NAND gate, acts as an inverter to obtain the normal AND output. For the OR operation, the normal inputs A and B are first complemented using two single input NAND gates. Now the complemented variables are fed as input to another NAND gate, which produces the normal ORed output. The implementation of Boolean functions with NAND gates may be obtained by means of a simple block diagram manipulation technique. The method requires that two other logic diagrams be drawn prior to obtaining the NAND logic diagram. The following steps are to be carried out in sequence: Step 1: From the given algebraic expression, draw the logic diagram with AND, OR, and NOT gates. Assume that both the normal (A) and complement (A) inputs are available. Step 2: Draw a second logic diagram with the equivalent NAND logic substituted for each AND, OR, and NOT gate. Step 3: Remove any two cascaded inverters from the diagram since double inversion does not perform any logical function. Also remove inverters connected to single external inputs and complement the corresponding input variable. The new logic diagram so obtained is the required NAND gate implementation of the Boolean function. Example 6.13. Construct a logic circuit for the Boolean expression A • B + C • (A + B • D) using only NAND gates.
Solution: The AND/OR implementation for the given Boolean expression is drawn in Figure 6.29(a). Now each AND gate is substituted by a NAND gate followed by an inverter and each OR gate is substituted by two input inverters followed by a NAND gate. Thus each AND gate is substituted by two NAND gates and each OR gate is substituted by three NAND gates. The logic diagram so obtained is shown in Figure 6.29(b). Note that Figure 6.29(b) has seven inverters (single input NAND gates) and five two-input NAND gates. Each two-input NAND gate has a number inside the gate symbol for identification purpose. Pairs of inverters connected in cascade (from each AND box to each OR box) are removed since they form double inversion, which has no meaning. The inverter connected to input A is removed and the input variable is changed from A to A . The result is the NAND logic diagram shown in Figure 6.29(c), with the number inside each NAND gate identifying the gate from Figure 6.29(b). This example demonstrates that the number of NAND gates required to implement the Boolean function is equal to the number of AND/OR gates, provided both the normal and complement inputs are available. Otherwise inverters must be used to generate any required complemented inputs. Example 6.14. Construct a logic circuit for the Boolean expression (A + E) • (B + C • D) using only NAND gates. Solution: The AND/OR implementation for the given Boolean expression is drawn in Figure 6.30(a). Now the NAND equivalent of each AND and each OR gate is substituted resulting in Figure 6.30(b). Note that Figure 6.30(b) has six inverters (single input NAND gates) and four two-input NAND gates. One pair of cascaded inverters may be removed. Also the three external inputs A, B, and E, which go directly to inverters, are complemented and the corresponding inverters are removed. The final NAND gate implementation so obtained is shown in Figure 6.30(c). The number inside each NAND gate of Figure 6.30(c) corresponds to the NAND gate of Figure 6.30(b) having the same number. For this example, the number of NAND gates required is equal to the number of AND/OR gates plus an additional inverter at the output (NAND gate number 5). In general, the number of NAND gates required to implement a Boolean function equals the number of AND/OR gates, except for an occasional inverter. This is true only when both normal and complemented inputs are available because the conversion forces certain input variables to be complemented. The Universal NOR Gate The NOR function is the dual of the NAND function. For this reason, all procedures and rules for NOR logic form a dual of the corresponding procedures and rules developed from NAND logic. Like the NAND gate; the NOR gate is also universal because it is alone sufficient to implement any Boolean function.
To show that any Boolean function can be implemented with the sole use of NOR gates, we need to only show that the logical operations AND, OR, and NOT can be implemented with NOR gates. This is shown in Figure 6.31 below. The NOT operation is obtained from a one-input NOR gate. Hence, a single input NOR gate is yet another inverter circuit. The OR operation requires two NOR gates. The first one produces the inverted OR and the second one being a single input NOT gate, acts as an inverter to obtain the normal OR output. The AND operation is achieved through a NOR gate with additional inverters in each input. Similar to the NAND logic diagram, the implementation of Boolean functions with NOR gates may be obtained by carrying out the following steps in sequence: Step 1: For the given algebraic expression, draw the logic diagram with AND, OR, and NOT gates. Assume that both the normal (A) and complement (Aj inputs are available. Steo 2: Draw a second logic diagram with equivalent NOR logic substituted for each AND, OR, and NOT gate. Step 3: Remove any two cascaded inverters from the diagram since double inversion does not perform any logical function. Also remove inverters connected to single external inputs and complement the corresponding input variable. The new logic diagram so obtained is the required NOR gate implementation of the given Boolean function. Example 6.15. Construct a logic diagram for the Boolean expression A • B + C • (A + B • D) using only NOR gates. Solution: The AND/OR implementation for the given Boolean expression is shown in Figure 6.32(a). Now each OR gate is substituted by a NOR gate followed by an inverter and each AND gate is substituted by two input inverters followed by a NOR gate. Thus each OR gate is substituted by two NOR gates and each AND gate is substituted by three NOR gates. The logic diagram so obtained is shown in Figure 6.32(b). Note that Figure 6.32(b) has eight inverters (single input NOR gates) and five two-input NOR gates. One pair of cascaded inverters (from the OR box to the AND box) may be removed. Also the five external inputs A, B, B, D and C, which go directly to inverters, are complemented and the corresponding inverters are removed. The final NOR gate implementation so obtained is shown in Figure 6.32(c). The number inside each NOR gate of Figure 6.32(c) corresponds to the NOR gate of Figure 6.32(b) having the same number. The number of NOR gates in this example equals the number of AND/OR gates plus an additional inverter in the-output (NOR gate number 6). In general, the number of NOR
gates required to implement a Boolean function equals the number of AND/OR gates, except for an occasional inverter. This is true only if both normal and complement inputs are available because the conversion forces certain input variables to be complemented. Combinational circuits are more frequently constructed with NAND or NOR gates than with AND, OR and NOT gates. NAND and NOR gates are more popular than the AND and OR gates because NAND and NOR gates are easily constructed with transistor circuits and Boolean functions can be easily implemented with them. Moreover, NAND and NOR gates are superior to AND and OR gates from the hardware point of view, as they supply outputs that maintain the signal value without loss of amplitude. OR and AND gates sometimes need amplitude restoration after the signal travels through a few levels of gates. Exclusive-OR and Equivalence Functions Exclusive-OR and equivalence, denoted by (c) and (c) respectively, are binary operations that perform the following Boolean functions: A+B=A•B+A•B A•B=A•B+A•B The truth table and the block diagram symbol for the exclusive-OR and the equivalence operations are shown in Figure 6.33 and Figure 6.34 respectively. Observe that the two operations are the complement of each other. Each is commutative and associative. Because of these two properties, a function of three or more variables can be expressed without parentheses as follows: Inputs Output A B C=A©B 0 0 0 0 1 1 1 0 1 1 1 0 (b) Truth table for Exclusive-OR. Figure 6.33. Block diagram symbols and truth table for Exclusive-OR operation. A a C = AOB 0 0 1 0 l 0 1 0 0 1 l 1 The exclusive-OR and equivalence operations have many excellent characteristics as candidates for logic gates but are expensive to construct with physical components. They are available as standard logic gates in IC packages but are usually constructed internally with other standard gates. For example, Figure 6.3 5(a) shows the implementation of a two-input exclusive-OR function with AND, OR and NOT gates. Figure 6.35(b) shows its implementation with NAND gates.
Only a limited number of Boolean functions can be expressed exclusively in terms of exclusive-OR or equivalence operations. Nevertheless, these functions emerge quite often during the design of digital systems. The two functions are particularly useful in arithmetic operations and error detection and correction. DESIGN OF COMBINATIONAL CIRCUITS The design of combinational circuits starts from the verbal outline of the problem and ends in a logic circuit diagram. The procedure involves the following steps: 1. State the given problem completely and exactly. 2. Interpret the problem and determine the available input variables and required output variables. 3. Assign a letter symbol to each input and output variables. 4. Design the truth table that defines the required relations between inputs and outputs. 5. Obtain the simplified Boolean function for each output. 6. Draw the logic circuit diagram to implement the Boolean function. The design procedure is illustrated below with the design of adder circuits because addition is the most basic arithmetic operation for any computer system. Addition in binary system can be summarized by the following four rules: 0+0=0 0+1=1 1 + 0 =1 1 + 1 = 10 The first three operations produce a sum whose length is one digit, but when both augend and addend bits are equal to 1, the binary sum consists of two digits. The higher significant bit of this result is called a carry. When the augend and addend numbers contain more significant digits, the carry obtained from the addition of two bits is added to the next higher order pair of significant bits. A combinational circuit that performs the addition of two bits is called a half-adder. One that performs the addition of three bits (two significant bits and previous carry) is called a full-adder. The name of the former stems from the fact that two half-adders can be employed to implement a full-adder. Design of Half-Adder From the definition of a half-adder, we find that this circuit needs two binary inputs and two binary outputs. The input variables designate the augend and addend bits whereas the output variables produce the sum and carry bits. Let A and B be the two inputs and S (for sum) and C (for carry) be the two outputs. The truth table of Figure 6.36 exactly defines the function of the half-adder. Inputs Outputs A 0 0 1
B 0 1 0
C 0 0 0
•s 0 1 1
1 1 1 Figure 6.36. Truth table for a half-adder.
0
The simplified Boolean functions for the two outputs, directly obtained from the truth table, are: The logic circuit diagram to implement this is shown in Figure 6.37. The half-adder is limited in the sense that it can add only two single bits. Although it generates a carry for the next higher pair of significant bits, it cannot accept a carry generated from the previous pair of lower significant bits. A full-adder solves this problem. Design of Full-Adder A full-adder forms the arithmetic sum of three input bits. Hence it consists of three inputs and two outputs. Two of the input variables (A and B) represent the augend and the addend bits and the third input variable (D) represents the carry from the previous lower significant position. Two outputs are necessary because the sum of three binary digits ranges in value from 0 to 3, and binary 2 and 3 need two digits. These two outputs are designated by the symbols S (for sum) and C (for carry). The truth-table of Figure 6.38 exactly defines the function of full-adder. The l's and O's for the output variables are determined from the arithmetic sum of the three input variables. When all input variables are 0, the output is 0 for both C and S. The S output is equal to 1 when only one input is equal to 1 or when all three inputs are equal to 1. The C output is 1 if two or three inputs are equal to 1. Inputs Outputs A B D C s 0 0 0 0 0 0 0 1 0 1 0 1 0 0 1 0 1 1 1 0 1 0 0 0 1 1 0 1 1 0 1 1 0 1 0 1 1 1 1 1 Figure 6.38. Truth table for a full-adder. The sum-of-products expressions for the two outputs can be directly obtained from the truth table and is given below: Although the expression fcjr S cannot be simplified, it is possible to simplify the expression for C as follows: The S output from the second half-adder is the exclusive-OR of D and the output of the first half-adder giving:
S=(A-B+A-B]-D+(A-B+A-B)-D
= AB AB D + ABD+ABD = A + B • A + B D+ABD+ABD =(A+B)-(A+B)-D+A-B-D+A-B-D = (AA+AB+AB + BB)-D+AB D+AB D =(AB+ABJD+ABD+AB D = ABD+A§;D+AB D+AB D = ABD+A-BD+ABD+ABD And we have the carry output C = lA-B + AB -D + A-B = ABD+ABD+A-B(D+D) This can be simplified as before to C = AB + AD + BD A Parallel Binary Adder Parallel binary adders are used to add two binary numbers. For example, if we want to add two four-bit numbers, we need to construct a parallel four-bit binary adder as shown in Figure 6.41. Such an adder requires one half-adder (denoted by HA) and three fulladders (denoted by FA). The binary numbers being added are A4 A3 A2 A! and B4 B3 B2 Bi and the answer is: A4 A3 A2 A1, + B4 B3 B2B1, S5 S4 S3 S2 S1 The first column requires only a half-adder. For any column above the first, there may be a carry from the preceding column. Therefore, we must use a full adder for each column above the first. To illustrate how the adder of Figure 6.41 works, let us see how it will add two numbers say, 9 and 11. The binary equivalent of decimal 9 is 1001, and that of decimal 11 is 1011. Figure 6.42 shows the binary adder with these inputs. As shown in the figure, the half-adder adds 1 + 1 to give a sum of 0 and a carry 1. The carry goes into the first full-adder, which adds 0 + 1 + 1 to get a sum of 0 and a carry of 1. This carry goes into the next full-adder, which adds 0 + 0 + 1 to get a sum of 1 and a carry of 0. The last full-adder adds 1 + 1 + 0 to get a sum of 0 and a carry of 1. The final output of the system is 10100. The decimal equivalent of binary 10100 is 20 which is the correct decimal sum of 9 and 11. The parallel binary adder of Figure 6.41 has limited capacity. The largest binary numbers that can be added using it are 1111 and 1111. Hence its maximum capacity is: 15 +15 30
1111 + 1111 11110 In order to increase the capacity, more full-adders can be connected to the left end of the system. For instance, to add six bit numbers, two more full-adders must be connected and for adding eight bit numbers, four more full-adders must be connected to the left end of the full-adder of Figure 6.41. Points to Remember 1. Boolean algebra deals with the binary number system. That is, the variables used in the Boolean equations may have only two possible values (0 and 1). 2. In Boolean algebra, the 'OR' operator used for logical addition is represented by the symbol '+'; the 'AND' operator used for logical multiplication is represented by the symbol ' • '; and the 'NOT' operator used for complementation is represented by the symbol'-'. 3. As regards precedence of Boolean operators, 'NOT' operator takes precedence over 'AND' and 'OR' operators, and 'AND' operator takes precedence over 'OR' operator. 4. The postulates of Boolean algebra are (a) A = 0 if and only if A is not equal to 1 (b) A = 1 if and only if A is not equal to 0 (c) x + 0 = x (d) x • 1 = x (e) x + y = y + x (f) x-y = y-x (g) x + (y + z) = (x + y) + z (h)x(yz) = (xy)-z (i) x • (y + z) = x • y + x • z (j) x + y-z = (x + y)-(x + z) (k) x + x = 1 (1) x • x = 0 5. In Boolean algebra, there is a precise duality between the operators • (AND) and + (OR) and the digits 0 and 1. This property is known as the principle of duality. 6. The basic Boolean identities are summarized in Figure 6.12. They are very helpful while dealing with Boolean algebra. 7. A Boolean function is an expression formed with binary variables, the two binary operators OR and AND, the unary operator NOT, parentheses and equal sign. 8. The complement of a Boolean function F is F and is obtained by interchanging O's for l's and l's for O's in the truth table that defines the function. 9. A sum-of-products expression is a product term (minterm) or several product terms (minterms) logically added (ORed) together. For example, the expression x • y + x • y is a sum-of-products expression.
10. A product-of-sums expression is a sum term (maxterm) or several sum terms (maxterms) logically multiplied (ANDed) together. For example, the expression (x + y) • (x + y) is a product of sums expression. 11. The sum-of-products and the product-of-sums form of Boolean expressions are known as standard forms. One prime reason for liking the sum-of-products or the product-of-sums expressions is their straightforward conversion to very nice gating networks. 12. A logic gate is an electronic circuit which operates on one or more input signals to produce standard output signals. These logic gates are the building blocks of all the circuits in a computer. 13. An AND gate is the physical realization of the logical multiplication (AND) operation. That is, it is an electronic circuit that generates an output signal of 1 only if all input signals are also 1. 14. An OR gate is the physical realization of the logical addition (OR) operation. That is, it is an electronic circuit that generates an output signal of 1 if any of the input signals is also 1. 15. A NOT gate is the physical realization of the complementation operation. That is, it is an electronic circuit that generates an output signal which is the reverse of the input signal. A NOT gate is also known as an inverter because it inverts the input. 16. A NAND gate is a complemented AND gate. That is, the output of NAND gate will be a 1 if any one of the inputs is a 0 and will be a 0 only when all the inputs are 1. 17. A NOR gate is a complemented OR gate. That is, the output of a NOR gate will be a 1 only when all inputs are 0 and it will be a 0 if any input represents a 1. 18. The logic gates are interconnected to form gating, or logic, networks which are known as combinational logic circuits. 19. The AND, OR, and NOT gates are logically complete in the sense that any Boolean function may be realizec using these three gates. 20. The NAND gate is said to be an universal gate because it is alone sufficient to implement any Booleai function. Similarly, the NOR gate is also an universal gate. 21. Exclusive-OR and equivalence, denoted by © and O respectively, are binary operations that perform the following Boolean functions: A©B=AB+AB 22. The design of combinational circuits starts from the verbal outline of the problem and ends in a logic circuit diagram. The procedure involves the following steps: (a) State the given problem completely and exactly. (b) Interpret the problem and determine the available input variables and required output variables. (c) Assign a letter symbol to each input and output variables. (d) Design the truth table that defines the required relations between inputs and outputs. (e) Obtain the simplified Boolean function for each output. (f) Draw the logic circuit diagram to implement the Boolean function. Questions 1. Explain the principle of duality in Boolean algebra. How is it useful? 2. Give the dual of the following Boolean expressions: (a) A + B (b)A + B + C
(c) A •B + A-B (d) A + B (e) A • (A + B) (f) A + A • B 3. Give the dual of the rule A+A •B = A + B 4. Prepare truth tables for the following Boolean expressions: (a) AB + AB (e) ABC (b) ABC + BC (f) AB • C + AB • C (c) A + B (g) (A + B) • (A + C) • (B (d) A + B + C (h) A • C + A •C 5. State and prove the two basic De Morgan's theorems. 6. Prove the following rules by the method of perfect induction: 7. Simplify the following Boolean expressions and draw logic circuit diagrams for your simplified expressions using AND, OR and NOT gates: 13. "AND, OR and NOT gates are logically complete". Discuss. 14. Why are NAND and NOR gates called universal122 I Foundations of Computing 15. Show the implementation of the logical operations AND, OR and NOT with only NAND gates and with only NOR gates. 16. Construct logic circuit diagrams for the Boolean expressions of Question 12 using only NAND gates. 17. Construct logic circuit diagrams for the Boolean expressions of Question 12 using only NOR gates. 18. Construct logic circuit diagram for a half-adder using only NAND gates. 19. Construct logic citcuit diagram for a half-adder using only NOR gates. 20. Why are combinational circuits more frequently constructed with NAND or NOR gates than with AND, OR and NOT gates? 22. Construct a logic circuit diagram for the exclusive-OR function using only NOR gates. 23. Construct a logic circuit diagram for the equivalence function using only NAND gates. 24. A logic circuit has three inputs, A, B and C. It generates an output of 1 only when A = 0, B = 1, C = 0 or when A = 1, B = 1, C = 0. Design a combinational circuit for this system. 25. A logic circuit has three inputs, A, B and C. It generates an output of 1 only under the following conditions: A = 0, B = 0, C = 0 A = 0,B=l,C=l A=l,B = 0,C=l A=1,B=1,C=1 Design a combinational circuit for this system. 26. Design a gating network which will have outputs 0 only when A = 0, B = 0, C = 0; A = 1, B = 0, C = 0;A = 1, B = 1, C = 0. The outputs are to be 1 for all other cases. 27. A three bit message is to be transmitted with an odd parity. An odd parity generator generates a parity bit (say P) so as to make the total number of Is odd (including P). That
is, P = 1 only when the number of Is in the input string is even. Design a combinational logic circuit for such a parity generator. 28. Design a combinational logic circuit to generate an even parity for hexadecimal digits.
Chapter 7 Processor and Memory We saw in Chapter 2 that the main components of a computer system are input, output, storage, and CPU. We also saw that storage units are basically of two types - primary and secondary. With this background knowledge, I we are now ready to learn more about the internal structure and working of the various components of a computer system. This and the next two chapters have been devoted to this learning. In this chapter we will learn about the structure and working of the CPU (also known as processor) and the primary storage (also known as main memory or simply memory). Subsequently, the structure and working of secondary storage devices and [input/output devices will be presented in Chapters 8 and 9 respectively. This chapter introduces the basic concepts related to the processor and the main memory of a computer system. In this chapter you will learn about: 1. How processors are internally structured? 2. How memory is structured? 3. How to determine the speed of a processor? 4. What are the different types of processors available? 5. How to determine the capacity of a memory? 6. What are the different types of memory available? 7. What are the main factors related to processor and memory that affect the processing speed of a computer system? 8. And several other terms related to the processor The basic processor and memory architecture of a computer system is shown in Figure 7.1. The details of the various components of this figure will be presented in this chapter. ROM PROM Flash Main Memory (RAM) V Cache Memory -----p. Decoder Accumulator register Program register
control
Instruction register Memory
General-purpose register General-purpose register •••
address
register Memory register
buffer
Input/Output register General-purpose register
General-purpose register
Control Unit Arithmetic Logic Unit Central Processing Unit I/O DEVICES Figure 7.1. Processor and memory architecture of a computer system. THE CENTRAL PROCESSING UNIT (CPU) The CPU is the brain of a computer system. All major calculations and comparisons performed by a computer are carried out inside its CPU. The CPU is also responsible for activating and controlling the operations of other units of the computer system. Hence no other single component of a computer determines its overall performance as much as the CPU. In order to be able to quickly evaluate any computer's capabilities, it is important to know how CPUs are internally structured, how different CPUs differ from each other, and how CPU speed is evaluated. These and other related concepts about CPU are described below in subsequent subsections. The Control Unit We saw in Chapter 2 (see Figure 2.1) that the two basic components of a CPU are the control unit and the arithmetic logic unit. The control unit of the CPU selects and interprets program instructions and then sees that they are executed. As shown in Figure 7.1, it has some special purpose registers (whose functions are described in a later subsection) and a decoder to perform these activities. The special purpose registers, namely the instruction register and the program control register respectively hold the current instruction and the next instruction to be executed and in this way help the control unit in instruction selection. On the other hand, the decoder has the necessary circuitry to decode and interpret the meaning of each and every instruction supported by the CPU. Each instruction is accompanied by microcode - very basic directions that tell the CPU how to execute the instruction. Although, the control unit does not perform any actual processing on the data, it acts as a central nervous system for the other components of the computer. It manages and coordinates the entire computer system including the input and output units. It obtains
instructions from the program stored in the main memory, interprets the instructions, and issues signals that cause other units of the system to execute them. The Arithmetic Logic Unit (ALU) The ALU of the CPU is the place where the actual execution of the instructions takes place during the data processing operation. That is, when the control unit encounters an instruction that involves an arithmetic operation (such as, add, subtract, multiply, divide) or a logic operation (such as less than, equal to, greater than), it passes control to the ALU. As shown in Figure 7.1, the ALU has some special purpose registers (whose functions are described in a later subsection) and the necessary circuitry to carry out all the arithmetic and logic operations included in the instructions supported by the CPU. For example, the control unit might load two numbers into the registers in the ALU. Then, it might tell the ALU to add the two numbers (an arithmetic operation), or to check if the two numbers are equal (a logical operation). In case of a microcomputer, the entire CPU (both the control unit and the ALU) is contained on a single tiny silicon chip called a microprocessor. Instruction Set Every CPU has the built-in ability to execute a particular set of machine instructions, called its instruction set. Most CPUs have 200 or more instructions (such as add, subtract, and compare) in their instruction set. The machine language designed for a specific CPU is based on the list of specific instructions supported by the CPU in its instruction set. Because each processor has a unique instruction set, machine language programs written for one computer will generally not run on another computer with a different CPU. CPUs made by different manufacturers have different instruction sets. In fact, different CPU models of the same manufacturer also often have different instruction sets. However, manufacturers tend to group their CPUs into "families" that have similar instruction sets. When a new CPU is developed, it is ensured that its instruction set includes all the instructions in the instruction set of its predecessor CPU, plus some new ones. This manufacturing strategy is known as upward compatibility, and the new CPU is said to be upward compatible with its predecessor. This feature allows software written for a computer with a particular CPU to work on computers with newer processors of the same family. In turn, it allows the users of these computer systems to easily upgrade their system without worrying about converting all their existing software. Registers As the instructions are interpreted and executed by the CPU, there is a movement of information between the various units of the computer system. In order to handle this process satisfactorily and to speed up the rate of information transfer, the computer uses a number of special memory units called registers. These registers are used to hold information on a temporary basis and are part of the CPU (not main memory).
The length of a register equals the number of bits it can store. Thus a register that can store 8 bits is normally referred to as an 8-bit register. Most CPUs sold today have 32-bit or 64-bit registers. The size of the registers, which is sometimes called the word size, indicates the amount of data, which the computer can process in a given period. The bigger the word size, the faster the computer can process a set of data. With all other parameters being same, a CPU with 32-bit registers can process data twice as fast as one with 16-bit registers. The number of registers varies among computers as does the data-flow pattern. Most computers use several types of registers, each designed to perform a specific function. Each of these registers possesses the ability to receive information, to hold it temporarily, and to pass it on as directed by the control unit. Although the number of registers varies from computer to computer, there are some registers that are common to all computers. The function of these registers is described below. 1. Memory Address Register (MAR). It holds the address of the active memory location. It is loaded from the program control register when an instruction is read from memory. 2. Memory Buffer Register (MBR). It holds the contents of the memory word read from, or written in, memory. An instruction word placed in this register is transferred to the instruction register. A data word placed in this register is accessible for operation with the accumulator register or for transfer to-the I/O register. A word to be stored in a memory location must first be transferred to the MBR, from where it is written in memory. 3. Program Control Register (PC). It holds the address of the next instruction to be executed. This register goes through a step-by-step counting sequence and causes the computer to read successive instructions previously stored in memory. It is assumed that instruction words are stored in consecutive memory locations and read and executed in sequence unless a branch instruction is encountered. A branch instruction is an operation that calls for a transfer to a non-consecutive instruction. The address part of a branch instruction is transferred to the PC register to become the address of the next instruction. To read an instruction, the contents of the PC register are transferred to the MAR and a memory read cycle is initiated. The instruction placed in the MBR is then transferred to the instruction register. 4. Accumulator Register (A). This register holds the initial data to be operated upon, the intermediate results, and also the final results of processing operations. It is used during the execution of most instructions. The results of arithmetic operations are returned to the accumulator register for transfer to main memory through the memory buffer register. In many computers there are more than one accumulator registers. 5. Instruction Register (I). It holds the current instruction that is being executed. As soon as the instruction is stored in this register, the operation part and the address part of the instruction (see Chapter 12) are separated. The address part of the instruction is sent to the MAR while its operation part is sent to the control section where it is decoded and interpreted_and ultimately command signals are generated to carry out the task specified by the instruction.
6. Input/Output Register (I/O). This register is used to communicate with the input/output devices. All input information such as instructions and data are transferred to this register by an input device. Similarly, all output information to be transferred to an output device is found in this register. Figure 7.2 summarizes the functions of each of these registers. SI. Name of Register Function No. 1 Memory Address (MAR) Holds the address of the active memory location 2 Memory Buffer (MBR) Holds information on its way to and from memory 3 Program Control (PC) Holds the address of the next instruction to be executed 4 Accumulator (A) Accumulates results and data to be operated upon 5 Instruction (I) Holds an instruction while it is being executed 6 Input/Output (I/O) Communicates with the I/O devices Figure 7.2. Functions of various registers. The execution of an instruction by the CPU during program execution normally involves the following steps: 1. The control unit takes the address of the next program instruction to be executed from the program control register and reads the instruction from the corresponding memory address into the instruction register of the control unit. 2. The control unit then sends the operation part and the address part of the instruction to the decoder and the memory address register respectively. 3. The decoder interprets the instruction and accordingly the control unit sends signals to the appropriate unit that needs to be involved in carrying out the task specified in the instruction. For example, if it is an arithmetic or logic operation, the signal is sent to the ALU. In this case the control unit also ensures that the data corresponding to the address part of the instruction is loaded in a suitable register in the ALU before the signal is sent to the ALU. The ALU performs the necessary operation on the data and signals the control unit as soon as it has finished. 4. As each instruction is executed, the address of the next instruction to be executed automatically gets loaded into the program control register, and steps 1 to 4 get repeated. Processor Speed The operations of the ALU and the control unit are performed with incredible speed. These operations are usually synchronized by a built-in electronic clock (known as system clock) that emits millions of regularly spaced electric pulses per second (known as clock cycles). Commands are interpreted and then executed at proper intervals, and the intervals are timed by a specific number of clock cycles. One cycle is the time it takes to perform one operation, such as moving a byte of data from one memory location to another. Normally several clock cycles are required to fetch, decode, and execute a single program instruction. The shorter the clock cycle, the faster the processor. Thus, the speed
with which an instruction is executed is directly related to the computer's built-in clock speed, which is the number of pulses, produced each second. This clock speed is measured in megahertz (or MHz), where mega means million and hertz means cycles per second. Over the years, clock speeds of processors have increased steadily. Most of today's popular personal computers have clock speeds in the 100- to 1000-MHz range. But microprocessors are now being built with even faster ratings. With all other parameters being equal (although they never are), a CPU operating at 500 MHz can process data 5 times as fast as one operating at 100 MHz. The number of operations performed per clock cycle also affects CPU performance. Most CPUs perform one operation per cycle. However, the new Pentium processors use a superscalar architecture that permits more than one instruction to be executed each clock cycle. This is achieved by having multiple ALUs on a single CPU chip so that two or more instructions can be executed simultaneously. Although the processing speed of personal computers is generally measured in MHz, the processing speed of workstations, mini computers, and mainframe systems is often measured in MIPS (Millions of Instructions Per Second) or BIPS (Billions of Instructions Per Second), and that of supercomputers is measured in MFLOPS (Millions of Floating Point Operations Per Second) or GFLOPS (gigaflops which refers to a billion FLOPS). This is because personal computers generally employ a single microprocessor chip as their CPU, but the other classes of computers often employ multiple processors to speed up their overall performance. Thus a workstation, minicomputer, or mainframe having a speed of 500-MIPS can execute 500 million instructions per second. Similarly, a supercomputer having a speed of 100-GFLOPS.can execute 100 billion floating point operations per second. The speed of supercomputers is measured in terms of FLOPS because supercomputer applications, which are often scientific, frequently involve floating point operations. Floating point operations deal with very small or very large numbers, which require very high precision data processing. CISC and RISC Processors The CPUs of earlier computers had small instruction sets, as it was difficult and expensive to build complex hardware circuits. However, the scenario changed with the advent of integrated circuits when it became inexpensive and easy to build complex hardware circuits. One of the earlier goals of CPU designers was to provide more and more instructions in the instruction set of a CPU to ensure that the CPU directly supports more features making it easier to translate high-level language programs to machine language and to ensure that the machine language programs run more effectively. Of course, every additional instruction in the instruction set of a CPU requires the necessary hardware circuitry to handle that instruction, adding more complexity to the CPU's hardware circuitry. Another goal of CPU designers was to optimize the usage of expensive memory. To achieve this, the designers tried to pack more instructions in memory by introducing the concept of variable length instructions such as half word, one and half word, etc. For example, an operand in an immediate instruction needs fewer bits and can be designed as a half word instruction. Additionally, CPUs were designed to
support a variety of addressing modes (discussed later in this chapter during the discussion of memory). CPUs with large instruction set, variable-length instructions, and a variety of addressing modes are said to employ CISC (Complex Instruction Set Computer) architecture. Since CISC processors possess so many processing features, they make the job of machine language programmers easier. However, they are complex and expensive to produce. Most of the personal computers of today use CISC processors. In early 1980s, some CPU designers discovered that several of the instructions supported by a CISC-based CPU are rarely used. Hence they came out with an idea that the complexity of CPU design can be greatly reduced by implementing only a bare minimum basic set of instructions plus some of the more frequently used instructions in the hardware circuitry of the CPU. Other complex instructions need not be supported in the instruction set of the CPU because they can always be implemented in software by using the basic set of instructions. While working on simpler CPU design, these designers also came up with the idea of making all the instructions of uniform length so that the decoding and execution of all instructions becomes simple and fast. Furthermore, to speed up computation and to reduce the complexity of handling a number of addressing modes, they decided to design al the instructions in such a way that they retrieve operands stored in registers in CPU rather than from memory These design ideas resulted in producing faster and less expensive processors. CPUs with a small instruction set fixed-length instructions, and reduced references to memory to retrieve operands are said to employ RISC (Reduced Instruction Set Computer) architecture. Since RISC processors have a small instruction set, they place extra demand on programmers, who must consider how to implement complex computations by combining simple instructions. However, due to simpler design, RISC processors are faster for most applications, less complex, and less expensive to produce than CISC processors. RISC processors are used extensively in medium size computers such as the IBM RS/6000 and high-end unix workstations. Popular RISC processors used in workstations are ALPHA (used in DEC-ALPHA workstations), RIOS (used in IBM workstations), SPARC used in SUN workstations), and PA-RISC (used in HP workstations). RISC processors are also used extensively in printers and other intelligent devices that have their own internal processors. Whether CISC or RISC technology will be the basis of most future processors has yet to be determined. Supporters of RISC technology claim that the limitations of a reduced instruction set are easily offset by the increased processing speed and the lower cost of RISC processors. But critics of RISC technology are of the opinion that more of these simple programmed instructions must be processed to complete a task, and an additional burden is thus placed on system software. The debate is still on and there seems to be no clear cut answer as to which technology is better. The answer may be that each technology lends itself best to certain applications, and so both technologies will coexist and be used where they are most efficient. EPIC Processors
The Explicitly Parallel Instruction Computing (EPIC) technology is fast emerging as the technology for next-generation processors. Hewlett-Packard and Intel Corporation have already adopted it in defining their new 64-bit Instruction Set Architecture (ISA), popularly known as IA-64. Processors based on IA-64 design will be simpler and more powerful than traditional CISC or RISC processors. These processors are mainly targeted to next-generation 64-bit high-end server and workstation market (not for personal computer market). The first processor based on IA-64, code-named Itanium (previously Merced), is an Intel product. It is believed that IA-64 based processors will surpass the performance of today's conventional RISC processors, but will protect customers' software investments by remaining backward compatible with IA-32 based processors (processors based on 32-bit Instruction Set Architecture). The three key features of EPIC technology targeted towards improved performance as compared to today's processors are explicit parallelism, predication and speculation. These terminologies are explained below. 1. Explicit Parallelism. EPIC technology breaks through the sequential nature of today's conventional processor architectures by allowing the software to communicate explicitly to the processor when operations can be done in parallel. For this it uses tighter coupling between the compiler and the processor, and enables the compiler to extract maximum parallelism in the original code and explicitly describe it to the processor. For example in case of IA-64, all instructions are of 32-bit fixed length. At compile time, the compiler detects which of the instructions can be executed in parallel. It then reorders them and groups them in such a way that instructions belonging to separate groups can be executed in parallel. At runtime, this explicit parallelism information provided by the compiler is exploited by the processor to execute the instructions faster. 2. Predication. Predication is a technique for obtaining improved performance by reducing the number of branches and branch mispredicts. Once again the help of compiler is first taken to reorder the instructions to reduce the number of branches as much as possible at compile time. The conventional processors use "branch prediction" technique in which the processor predicts which way a branch will fork and speculatively executes instructions along the predicted path. At the time of execution of the branch instruction, if the prediction is found to be found to be correct, the processor gains performance improvement because the instructions lying in the path to be executed now have already been executed and their results can be directly used now by the processor. However, if the prediction is found to be wrong, the results of execution of the predicted path are discarded and the instructions of the correct path are now taken up for execution. However, the EPIC technology uses "branch predication" instead of "branch prediction". In this technique, instead of predicting and executing one of the paths of a branch, the processor executes the instructions of all the paths of the branch exploiting as much parallelism as possible. Now when the processor discovers the actual branch outcome, it retains the valid results and discards other results. Thus branch predication effectively removes the negative impact of branch prediction technique in cases of branch4 mispredict.
3. Speculation. Speculation is a technique used to obtain improved performance by reducing the effect of memory-to-processor speed mismatch. The memory access speed being much slower than the processor speed, it is desirable that data required during the course of execution is loaded and kept ready before it is actually required by the processor. Speculative data loading technique is designed to take care of this requirement. Its main goal is to separate the loading of data from its actual use. It not only allows the processor to load a piece of data from memory before a program actually needs it, but it also postpones the reporting of exceptions if the loaded data is illegal. Speculation technique is also implemented by taking help of the compiler. For this, the compiler analyses the program at compile time, looking for any instructions that will need data from memory. It inserts speculative load instructions in the instructions stream of the program well ahead of the instructions that need data from memory. It also inserts a matching speculative check instruction immediately before the instructions that need data from memory. It now reorders the surrounding instructions so that the processor can dispatch them in parallel. Now when the processor encounters a speculative load instruction at runtime, it retrieves and loads the data from the memory. When the processor encounters the speculative check instruction, it verifies the load before allowing the program to use the loaded data in the next instruction. If the load is invalid, the processor does not immediately report an exception. It postpones the exception reporting until it encounters a check instruction that matches the speculative load. If the load is valid, the system behaves as if the exception never happened. Speculation technique combined with predication technique gives the compiler more flexibility to reorder instructions and increase parallelism. Notice that all the three techniques are based on the availability of an intelligent compiler and closed coupling between the compiler and the processor. Hence the performance of processors based on EPIC technology comes from both hardware and software. THE MAIN MEMORY We saw in the discussion above that the CPU contains the necessary circuitry for data processing and controlling the other components of a computer system. However, one thing it does not have built into it is the place to store programs and data that are needed during data processing. We also saw that the CPU does contain several registers for storing data and instructions, but these are very small areas that can hold only a few bytes at a time and are just sufficient to hold only one or two instructions and the corresponding data. If the instructions and data of a program being executed by the CPU were to reside in secondary storage like a disk and fetched and loaded one by one into the registers of the CPU as the program execution proceeded, this would lead to the CPU being idle most of the time because there is a large speed mismatch between the rate at which CPU can process data and the rate at which data can be transferred from disk to CPU registers. For example, a CPU can process data at a rate of about 5 nanosecond/byte and a disk reader can read data at a speed of around 5 microsecond/byte. Thus, within the time in which the
disk can supply one byte of data, the CPU can process 1000 bytes. This would lead to a very slow overall performance even if the computer system used a very fast CPU. To overcome this problem, there is a need to have a reasonably large storage space, which can hold the instructions and data of the program(s) the CPU is currently woi*king on? The time to fetch and load data from this storage space into the CPU registers must also be very small as compared to that for disk storage to reduce the speed mismatch problem with the CPU speed. Every computer has such a storage space, which is known as primary storage, main memory, or simply memory. It is a temporary storage area that is built into the computer hardware and in which instructions and data of a program reside mainly when the program is being executed by the CPU. Physically, this memory consists of some chips either on the motherboard or on a small circuit board attached to the motherboard of a computer system. This built-in memory allows the CPU to store and retrieve data very quickly. The rate of fetching data from this memory is typically of the order of 50 nanosecond/byte. Hence the rate of data fetching from the main memory is about 100 times faster than that from a high-speed secondary storage like disk. Storage Evaluation Criteria Any storage unit of a computer system is characterized and evaluated based on the following properties: 1. Storage capacity. It is the amount of data that can be stored in the storage unit. A large capacity is desired. As compared to secondary storage units, primary storage units have less storage capacity. 2. Access time. This is the time required to locate and retrieve stored data from the storage unit in response to a program instruction. A fast access time is preferred. As compared to secondary storage units, primary storage units have faster access time. 3. Cost per bit of storage. This refers to the cost of a storage unit for a given storage capacity. Obviously, a lower cost is desirable. As compared to secondary storage units, primary storage units have higher cost per bit of storage. 4. Volatile. If the storage unit can retain the data stored in it even when the power is turned off or interrupted, it is called non-volatile storage. On the other hand, if the data stored are lost when the power is turned off or interrupted, it is called volatile storage. Obviously, a non-volatile storage is desirable. In almost all computer systems, the primary storage units are volatile and the secondary storage units are non-volatile. 5. Random access. If the time taken to access a piece of data from the storage unit is independent of the location of the data in the storage unit, it is called a random access storage or random access memory (RAM). Each separate location of a RAM is as easy to access as any other location, and takes the same amount of time. In almost all computer systems, the primary storage units have random access property and the secondary storage units have either pseudo-random access (access time is nearly same for all locations but not exactly same) or sequential access (access time directly depends on the location of the data) property. Main Memory Organization
A primary storage or main memory of a computer system is made up of several small storage areas called locations or cells. Each of these locations can store a fixed number of bits called word length of that particular memory. Thus, as shown in Figure 7.3, a given memory is divided into N words, where N generally is some power of 2. Each word or "location has a built-in and unique number assigned to it. This number is called the address of the location and is used to identify the location. Each location can hold either a data item or an instruction, and its address remains the same regardless of its contents. The addresses normally start at 0 and the highest address equals the number of words that can be stored in the memory minus 1. For example, if a memory has 1024 locations, then th (c) address ranges between 0 and 1023. Thus, at address 0 we find a word, at address 1 a second word, at address 2 a third word, and so on up to the final word at the largest address. There is an important difference between the address number and the contents of the address. A memory is like a large cabinet containing as many drawers as there are addresses in memory. Each drawer contains a word and the address of each word is written on the outside of the drawer. If we write or store a word say 10101010 at address 125, it is like placing the word 10101010 in the drawer labeled 125. Later, reading from address 125 is like looking in that drawer to see its contents which is now 10101010. We do not remove the word at an address when we read, but change the contents at an address only when we store or write a new word. Thus, entering data into a storage location is destructive of previous contents, but retrieving data from a location is nondestructive. The act of entering data into a storage location is called memory write operation, and the act of retrieving data from a storage location is called memory read operation. Hence read is a non-destructive operation and write is a destructive operation. Why More Bits? You might have heard about 8-bit computers, 16-bit computers, 32-bit computers, etc. This refers to the word length of the memory of a particular computer in terms of total number of bits per memory word. Word length is an important architectural factor. Small machines have word lengths of 8, 16, or 32 bits; large machine word lengths are 64 bits or more. The obvious question that arises to ones mind is that what is the advantage of having more number of bits per word instead of having more words of smaller size (length)? For an answer to the above question, imagine a highway with eight lanes, and a heavy flow of traffic. If it is expanded to sixteen lanes, the flow of traffic speeds up considerably. "8 bits" refers to the number of "lanes" on a microchip. More bits means a more rapid flow of electronic signals. In other words, a faster computer. Thus, what an 8bit computer takes one miriute to db, a 32-bit computer may do in few seconds. Data and instructions are moved to and from memory in bunches of word length. Therefore, even if the electronic circuits used are comparable in speed, machines having smaller word-length will be slower in operation than machines having larger word-length. This difference is analogous to providing the user of a machine having smaller word
length with a small data shovel and the user of a machine having larger word length with a large data shovel. Even though they both may be shoveling at comparable speeds, the user with the smaller shovel will be slower because more shovelfuls are needed to move the same amount of data. Fixed and Variable Word-length Memory The main memory of some computers is designed to store a fixed number of characters (equal to its word-length in bytes) in each numbered address location. Such computers are said to be word-addressable, and they employ a fixed-word-length memory approach. In these computers, storage space is always allocated in multiples of word-length. So if a word addressable computer has a fixed word-length of 4 bytes (4 characters) then this computer will require one word (4 bytes) to store the word "CAT" and two words (8 bytes) to store the word "BOMBAY". In many computers, the main memory is also designed in such a way that each numbered address can only store a single character (A, B, 1,2, +, -, etc.). Computers designed in this manner are said to be character-addressable and they employ a variable-word-length memory approach. Thus in these machines, only 3 bytes will be required to store the word "CAT" and only 6 bytes will be required to store the word "BOMBAY". Figure 7.4 summarizes the difference between the fixed-word-length and variable-word-length memory approaches. Both the fixed and the variable word-length memory systems have their own merits and demerits. The fixed-word-length memory approach is normally used in large scientific computers for gaining speed of calculation. On the other hand, the variable-word-length approach is used in small business computers for optimizing the use of storage space. For example, let us consider a fixed-word-length machine with a word size of eight characters. If most of the data words to be stored are of less than five characters then more than half of the storage space will remain unused. This will not happen in case of a machine with variable-word-length memory because a character can be placed in every storage cell of this machine. However, word-addressable computers possess faster calculating capability because they can add two data words in a single operation. If the fixed-length word is eight characters, two eight digit numbers can be added in single operation. On the other hand, with a character-addressable machine only one digit in each number is added during a single operation and eight steps would thus be needed to add two eight digit numbers. A computer may be designed to employ either a fixed-word-length or a variable-wordlength memory organization. The set of instructions available with these computers allow them to be operated as either variable-or fixed-word-length memory computers. However, with the memory becoming cheaper and larger day-by-day, most modern computers employ fixed-word-length memory organization.
Main Memory Capacity The main memory capacity of large computer systems is normally more than small systems. This capacity is defined in terms of the number of bytes a computer system can store. Memory capacity of a computer system is normally stated in terms of kilobytes (KB) which is equal to 1024 (2 ) bytes of storage, or in terms of megabytes (MB) which is equal to 1,048,576 (220) bytes of storage, or in terms of gigabytes (GB) which is equal to 1,073,741,824 (230) bytes of storage. Notice that 1 KB is about 103 bytes, 1 MB is about 106 bytes, and 1 GB is about 109 bytes, thus the origin of the prefixes kilo (thousand), mega (million) and giga (billion). Few years ago, memory capacities were stated only in terms of KB. With the rapidly increasing capacity of memory, today memory capacity of small computer systems is stated in terms of MB and that of large computer systems is stated in terms of GB. The memory capacity continues to increase rapidly and it is only a matter of time before we start stating memory capacity in terms of terabytes (TB), which is about one trillion (1012) bytes. Today, GB and TB are frequently used to refer to the capacity of secondary storage of computer systems. Thus a computer having 32 MB of memory is capable of storing 32 x 1048576 = 33,554,432 bytes or characters. Sometimes the memory capacity of a computer system is also stated in terms of number of words. However, if the memory capacity is stated in terms of words, it is necessary to know the word size in bits or bytes in order to determine the actual memory capacity of the computer. Thus, while specifying the memory capacity in terms of words, it is customary to specify the total number of bits per word along with the total number of words. So a memory with 4096 locations each with a different address and with each location storing 16 bits is called a "16-bit 4096-word memory, or, in the vernacular of the computer trade, a "4 K 16-bit memory". Similarly, a memory having 215 words with each word of 16 bits is called a "32 K 16-bit memory". If the word size of a memory is 8 bits (equal to a byte) then it becomes immaterial whether the memory capacity is expressed in terms of bytes or words. Thus a memory having 216 words with each word of 8 bits is simply referred to as 64 K memory (word size of 8-bits is implicit here). Static and Dynamic RAM When people talk about computer memory, they usually mean the volatile RAM memory. Physically, this memory consists of some integrated circuit chips (IC chips shown in Figure 1.2) either on the motherboard or on a small circuit board attached to the motherboard. A computer's motherboard is designed so that its memory capacity can be easily enhanced by adding more memory chips. Hence, if you decide to have more memory than your computer currently has, you can buy more memory chips, open up your computer, and plug it in the empty memory slots on the motherboard. This job is normally done by the service engineers. The additional RAM chips, which plug
into'special sockets on the motherboard are also known as single in-line memory modules {SIMMs). RAM chips are of two types - dynamic and static. A dynamic RAM (DRAM), uses an external circuitry to periodically "regenerate" or refresh the storage charge to retain the stored data. On the other hand, a static RAM (SRAM) does not need any special regenerator circuits to retain the stored data. Since it takes more transistors and other devices to store a bit in a static RAM, these chips are more complicated and take up more space for a given storage capacity than do dynamic RAMs. Moreover, a static RAM is faster, costlier, and consumes more power than dynamic RAM. Due to these reasons, large memories use dynamic RAM, and static RAM is used mainly for specialized applications. The main memory of most computers uses dynamic RAM. Read-Only Memory A special type of RAM, called read-only memory (ROM) is a non-volatile memory chip in which data is stored permanently and cannot be altered by the programmer. In fact, storing data permanently into this kind of memory is called "burning in the data", because data in such memory is stored by using fuse-links. Once a fuse-link is burnt it is permanent. The data stored in a ROM chip can only be read and used - they cannot be changed. This is the reason why it is called read-only memory (ROM). Since ROM chips are non-volatile, the data stored inside a ROM are not lost when the power supply is switched off, unlike the case of a volatile RAM chip. ROMs are also known as field stores, permanent stores, or dead stores. ROMs are mainly used to store programs and data which do not change and are frequently used. For example, the most basic computer operations are carried out by wired electronic circuits. However, there are several higher level operations that are very frequently used but will require very complicated electronic circuits for their implementations. Hence instead of building electronic circuits for these operations, special programs are written to perform these operations. These programs are called microprograms because they deal with low-level machine functions and are essentially substitutes for additional hardware. Microprograms are written to aid the control unit in directing all the operations of the computer system. ROMs are mainly used by computer manufacturers for storing these microprograms so that they cannot be modified by the users. A good example of a microprogram that is stored in a ROM chip of a computer is the set of instructions that is needed to make the computer system ready for use when its power is switched on. This microprogram called "system boot program", contains a set of startup instructions to check if the system hardware like memory, I/O devices, etc. are functioning properly, and looks for an operating system and loads its core part in the volatile RAM of the system to produce the initial display-screen prompt. Note that this microprogram is used every time the computer is switched on and needs to be retained when the computer is switched off. Hence ROM is an ideal storage for this microprogram.
There are two types of read-only memory (ROM) - manufacturer-programmed and userprogrammed. A manufacturer-programmed ROM is one in which data is burnt in by the manufacturer of the electronic equipment in which it is used. For example, a personal computer manufacturer may store the system boot program permanently in the ROM chip used on the motherboard of all the PCs manufactured by it. Similarly, a printer manufacturer may store the printer controller software in the ROM chip used on the circuit board of all the printers manufactured by it. Manufacturer-programmed ROMs are mainly used in those cases where the demand for such programmed ROMs is large. Note that manufacturer-programmed ROM chips are supplied by the manufacturers of electronic equipment, and it is not possible for a user to modify the programs or data stored inside the ROM chip. On the other hand, a user-programmed ROM is one in which the user can load and store "read-only" programs and data. That is, it is possible for a user to "customize" a system by converting his/her own programs to microprograms and storing them in a user-programmed ROM chip. Such a ROM is commonly known as Programmable Read-Only Memory (PROM) because a user can program it. Once the user programs are stored in a PROM chip, they can usually be executed in a fraction of the time previously required. PROMs are; programmed to record information using a special device known as PROM-programmer. However, once the chip has been programmed, the recorded information cannot be changed, i.e., the PROM becomes a ROM and it/Is only possible to read the stored information. PROM is also non-volatile storage, i.e., the stored information remains intact even if power is switched off. Once information is stored in a ROM chip or a PROM chip it cannot be altered. However, there is another type of memory chip called Erasable Programmable ReadOnly Memory (EPROM) that overcomes this problem. As the name implies, it is possible to erase information stored in an EPROM chip and the chip can be reprogrammed to store new information. EPROMs are often used by R&D personnel (experimenters) who frequently change the microprograms to test the efficiency of the computer system with new programs. EPROMs are also useful in case of those applications where one may like to store a program in a ROM, which would normally not change, but under some unforeseen conditions one may like to alter it. When an EPROM is in use, information stored in it can only be "read" and the information remains in the chip until it is erased. EPROM chips are of two types - one in which the stored information is erased by exposing the chip for some time to ultraviolet light, and the other one in which the stored information is erased by using high voltage electric pulses. The former is known as Ultra Voilet EPROM (UVEPROM) and the latter is known as Electrically EPROM (EEPROM). It is easier to alter information stored in an EEPROM chip as compared to an UVEPROM chip. This is because to alter information stored in an UVEPROM chip, the chip has to be removed from the computer, erased using a special ultra-violet rays emitting device, and then reprogrammed. On the other hand in case of an EEPROM chip, stored programs can be altered electrically by using special programs, without the chip being removed from the computer. Due to the ease with which stored programs can be altered, EEPROM is also known as flash memory. Flash memory is used in many new I/O and storage devices. The intelligence of these devices can be upgraded by simply down-loading new software from a vendor-supplied disk to flash memory. Upgrades to
early peripheral devices required the user to replace the old circuit board or chip with a new one. The various types of random-access memories are summarized in Figure 7.5. Regardless of the type cf ROM chip used, they all serve to increase the efficiency of a CPU by controlling the performance of a few specialized tasks. A generalized CPU can be made to meet the unique needs of different users merely by changing microprograms. Random Access Memory (RAM) Volatile and Writable Non-volatile and Read-Only Static (SRAM) Dynamic (DRAM) Manufacturer-programmed (ROM) User-programmed PROM EPROM UVEPROM EEPROM Minimizing Memory-Processor Speed Mismatch We saw that the use of main memory helps in minimizing the disk-processor speed mismatch to a large extent because the rate of data fetching by the CPU from the main memory is about 100 times faster than that from a high-speed secondary storage like disk. However even with the use of main memory, the memory-processor speed mismatch becomes a bottleneck in the speed with which the CPU can process instructions because there is a 1 to 10 speed mismatch between the processor and the memory. That is, the rate at which data can be fetched from memory is about 10 times slower than the rate at which CPU can process data. Hence in many situations the performance of the processors gets limited due to the slow speed of the main memory. Obviously, the overall performance of a processor can be greatly improved by minimizing the memory-processor speed mismatch. The two commonly used approaches for minimizing the memory-processor speed mismatch are the use of cache memory (pronounced "cash" memory) and memory interleaving. These techniques are described below. Cache Memory Cache memory is an extremely fast, small memory between CPU and main memory whose access time is closer to the processing speed of the CPU. It acts as a high-speed buffer between CPU and main memory and is used to temporarily store very active data and instructions during processing. Since the cache memory is faster than main memory, the processing speed is increased by making data and instructions needed in present processing available in the cache., The obvious question that arises is how can the system know in advance which data and instructions are needed in present processing so as to make it available beforehand in the
£ache. The answer to this question comes from a principle known as locality of reference. According to this principle, during the course of execution of most programs, memory references by the processor, for both instructions and data, tend to cluster. That is, if an instruction is executed, there is a likelihood of the same instruction being executed again soon. This is true because most programs typically contain a number of iterative loops (such as while and for loops). Once a loop is entered during the execution of a program, there are repeated references to a small set of instructions (i.e., the instructions in the loop are executed again and again several times). A branch instruction (such as a go to statement) upsets the locality of reference to instructions, but such instructions do not occur frequently in well written programs. Locality of reference is true not only for references to program instructions but also for references to data. For example, operations on tables, arrays, and records in a file involve access to a clustered set of data words. Over a long period, the instructions or data clusters in use change, but over a short period, the processor primarily work's with fixed clusters of instructions and data resulting in clustered memory references. Based on the feature of locality of reference, the effective use of a cache is made in the following manner. As shown in Figure 7.6, the cache memory acts as a small, fast-speed buffer between the processor and main memory. It contains a copy of a portion of main memory contents. When a program is running and the CPU attempts to read a word of memory (instruction or data), a check is made to determine if the word is in the cache. If so, the word is delivered to the CPU from the cache. If not, a block of main memory, consisting of some fixed number of words including the requested word, is read into the cache and then the requested word is delivered to the CPU. Because of the feature of locality of reference, when a block of memory words is fetched into the cache to satisfy a single memory reference, it is likely that there will soon be references to other words in that block. That is, the next time the CPU attempts to read a word, it is very likely that it finds it in the cache and saves the time needed to read the word from main memory. You might think that the odds of the CPU finding the word it needs in the cache are small, but it actually does find the word it needs there often enough to improve perceptibly the performance of a processor. In fact, it has been found in practice that more than 90% of time, the needed word is available in the cache. Main Memory (Slower in speed and larger in size than cache memory) Block of words including the accessed word Address of accessed word when the word is not found in the cache Cache Memory Accessed word Address of accessed word CPU As the name implies, cache memory is a memory in hiding (the word "cache" literally means a hiding place for treasure or stores) and is not addressable by the normal users of the computer system. Needed instructions/data are transferred automatically between the
cache and main memory by the hardware without any programmer intervention. In fact, the application programmers are unaware of its presence and use. Figure 7.6 illustrates the general concept of cache memory usage. Actual systems may be designed to have some variations. For example, many computer systems are designed to have two separate cache memories called instruction cache and data cache. The instruction cache is used for storing program instructions and the data cache is used for storing data. This allows faster identification of availability of accessed word in the cache memory and helps in further improving the processor speed. Many computer systems are also designed to have multiple levels of caches (such as level one and level two caches, often referred to as LI and L2 caches). LI cache is smaller than L2 cache and is used to store more frequently accessed instructions/data as compared to those in the L2 cache. The use of cache memory requires several design issues to be addressed. A detailed discussion of cache design is beyond the scope of this book. Key design issues are briefly summarized below. 1. Cache size. Cache memory is very expensive as compared to the main memory and hence its size is normally kept very small. It has been found through statistical studies that reasonably small caches can have a significant impact on processor performance. Hence cache memory size is usually decided based on statistical studies of program behavior and is chosen so that the probability of the needed instruction/data being in the cache is more than 90%. As a typical example of cache size, a system having 1 GB of main memory may have about 1 MB of cache memory. Many of today's personal computers have 64 KB, 128 KB, 256 KB, 512 KB, or 1 MB of cache memory. 2. Block size. Block size refers to the unit of data (few memory words) exchanged between cache and main memory. As the block size increases from very small to larger sizes, the hit ratio (fraction of time that referenced instruction/data is found in cache) will at first increase because of the principle of locality since more and more useful words are brought into the cache. However, the hit ratio will begin to decrease as the block size further increases because the probability of using the newly fetched words becomes less than the probability of reusing the words that must be moved out of the cache to make room for the new block. Based on this fact, the block size is suitably chosen to maximize the hit ratio. 3. Replacement policy. When a new block is to be fetched into the cache, another may have to be replaced to make room for the new block. The replacement policy decides which block to replace in such situation. Obviously, it will be best to replace a block that is least likely to be needed again in the near future. Although it is impossible to identify such a block, a reasonably effective strategy is to replace the block that has been in the cache longest with no reference to it. This policy is referred to as the least recently used (LRU) algorithm. Hardware mechanisms are needed to identify the least recently used block.
4. Write policy. If the contents of a block in the cache are altered, then it is necessary to write it back to main memory before replacing it. The write policy decides when the altered words of a block are written back to main memory. At one extreme, an updated word of a block is written to the main memory as soon as such updates occur in the block. At the other extreme, all updated words of the block are written to the main memory only when the block is replaced from the cache. The latter policy minimizes memory write operations but temporarily leaves main memory in an inconsistent (obsolete) state. This can interfere with multiple-processor operation when multiple processors share a common memory and cache. It can also interfere with direct memory access by input/output modules. Several other write policies that are somewhere in between these two extremes are also in use. The choice of a suitable write policy for a particular computer architecture is normally left to the designer. Memory Interleaving In this method, the main memory is divided into n equal size modules and the CPU has separate MAR and MBR registers for each memory module. In addition, the CPU has n instruction registers and a memory access system. When a program is loaded into the main memory, its successive instructions are stored in successive memory modules. For example, if n = 4 and the four memory modules are Mi, M2, M3 and M4, the 1st instruction will be stored in Mi, 2nd in M2, 3td in M3,"4th in M4, 5th in Mi, 6th in M2, and so on. Now during the execution of the program, when the processor issues a memory fetch command, the memory access system creates n consecutive memory addresses and places them in the n MARs in the right order. A memory read command reads all the n memory modules simultaneously, retrieves the n consecutive instructions, and loads them into the n instruction registers. Thus each fetch for a new instruction results in the loading of n consecutive instructions in the n instruction registers of the CPU. Since the instructions are normally executed in the sequence in which they are written, the availability of n successive instructions in the CPU avoids memory access after each instruction execution, and the total execution time speeds up. Obviously, the fetched successive instructions are not useful when a branch instruction is encountered during the course of execution. This is because the new set of n successive instructions starting from the location of the instruction to which the control branches will be loaded into the n instruction registers, overwriting the previously stored instructions which were loaded but some of which were not executed. However, the method is quite effective in minimizing the memory-processor speed mismatch because branch instructions do not occur frequently in a program. Hybrid Approach This method uses both cache memory and memory interleaving. Cache memory is used to reduce the speed mismatch between the CPU and the main memory, and memory interleaving is used to reduce the speed mismatch between the CPU and the cache memory. This approach further helps in speeding up the total execution time. MEMORYBUSES
We saw in the discussion above that the data and instructions of a program being executed are stored in the memory and are fetched and loaded into the CPU registers as the program execution proceeds. That is, the interaction between the CPU and memory takes place very frequently. To enable this interaction, some type of connectivity is needed between the two units of a computer system. This connectivity channel is known as a bus. Physically, a bus is a set of wires, which carries a group of bits in parallel and has an associated control scheme. The bus width is defined as the number of parallel lines (wires) in the bus. Every computer system has the following three types of buses for interconnecting the CPU and memory: 1. Data Bus. The data bus is used to transfer data between the CPU and memory. The one that you hear the most about is the data bus. Hence when people refer to the bus of a computer system, they usually mean the data bus. The bus width of the data bus is an important parameter that affects the overall speed of a computer system. This is because each wire of a bus can transfer one bit at a time. Hence an 8-bit bus (one having 8 parallel wires) can move 8 bits (one byte) at a time, a 16-bit bus can transfer two bytes, and a 32-bit bus can transfer four bytes at a time. This is similar to a multi-lane highway. The wider the highway, the more traffic can flow simultaneously. Similarly, a wider data bus enables more bits of data to travel simultaneously resulting in faster exchange of data. 2. Address Bus. We saw that every storage location in the memory has a unique address. The address of a location does not change, but the data stored in it can change. In order to retrieve some data from memory, it is necessary to specify the address of the location where the data is stored. The address bus is used to carry the address of a memory location whenever data is to be transferred to or from memory. Obviously, the width of the address bus must be equal to the number of bits in the memory address register (MAR) of the system. The width of the address bus is an important parameter for a computer system because it determines the maximum number of memory locations that the computer system can have. For example, if the address bus was only 8 bits wide, the CPU could address only 256 bytes of memory. Most of early personal computers had 20-bit wide address buses, so the CPU could address 2 20 bytes, or 1 MB, of memory. Today, most systems have 32-bit wide address buses that can address 232 bytes (over 4 billion bytes), or 4 GB, of memory. 3. Control Bus. In addition to sending addresses and exchanging data with the memory, the CPU also needs to send control signals to the memory to specify whether the data is to be read from or written to the specified address location. Such signals, which are in the form of READ/WRITE commands, are carried by the control bus. Thus a memory access by the CPU involves all the three buses - the memory access command is carried by the control bus, the memory address is carried by the address bus, and the data to be transferred is carried by the data bus. Figure 7.8 shows the three buses used for interconnecting the CPU and main memory for enabling this. CPU Data Bus Address Bus
Control Bus Main Memory Figure 7.8. The three types of buses used between the CPU and main memory. I/O BUSES We saw in Chapter 2 that in addition to the flow of data between the CPU and memory, data also flows between the CPU and I/O devices in a computer system. Hence just like the buses between the CPU and memory, every computer system also uses buses for interconnecting the CPU with I/O devices. The same three types of buses used for interconnecting the CPU and memory are also used for interconnecting the CPU with I/O devices. Their roles in context of I/O devices are described below. 1. Data Bus. The data bus is used to transfer data between the CPU and I/O devices. As discussed before, a wider data bus will enable faster exchange of data between the CPU and I/O devices. The commonly used, industry standard data buses are as follows: - ISA Bus. ISA stands for Industry Standard Architecture. It is a 16-bit bus but that can transmit data along either 8 or 16 data lines, depending on what kind of adapter card is used in an expansion slot. That is, ISA expansion slots can accept both 8-bit and 16-bit adapter cards. MCA Bus. MCA stands for Micro Channel Architecture. It is a 32-bit bus that transmits data along 32 data lines. Due to its wider bus width, it is significantly faster than the ISA bus. However, MCA expansion slots cannot accept 8-bit or 16-bit adapter cards and require specifically designed 32-bit adapter cards. EISA Bus. EISA stands for Extended Industry Standard Architecture. Since the expansion slots that fit into an ISA bus could not work with the MCA bus, the MCA bus had the problem of upward compatibility. EISA bus was designed to solve this problem. Hence like MCA, EISA is also a 32-bit bus. However, unlike MCA, it was designed to accept and use the old ISA expansion slots. The EISA is faster than the ISA, but not as fast as the MCA -the price of its compatibility with the older 16-bit expansion slots. 2. Address Bus. A computer system normally has multiple I/O devices like disk, tape, network, etc. simultaneously connected to it. Each I/O device has a unique identifier (or address) associated with it. The address bus is used to carry the address of the I/O device to be accessed by the CPU. 3. Control Bus. The control bus is used to carry commands such as START, READ, WRITE, REWIND TAPE, etc., from the CPU to I/O devices. It is also used to carry the status information of the I/O devices to the CPU. Thus an I/O device access by the CPU involves all the three buses - the device access command is carried by the control bus, the device address is carried by the address bus, and the data to be transferred is carried by the data bus. Figure 7.9 shows the three buses used for interconnecting the CPU and I/O devices for enabling this. Each device has a controller, which controls the operation of the device. It also temporarily stores the data transferred between the CPU and the device in a local data register. INTERCONNECTION ARCHITECTURES
In the discussion above, we saw the use of buses in interconnecting the CPU with memory and I/O devices. The interconnection architecture defines how exactly these functional units of a computer system are connected to each other. The accessing mechanism and the flow of data from one unit to another depends on the interconnection architecture. The three commonly used interconnection architectures are described below. Unibus Architecture As shown in Figure 7.10, in the unibus architecture there is a single data bus, a single address bus, and a single control bus used to interconnect the CPU with both the memory and I/O devices. That is, the buses are shared for data transfer between the CPU and memory, as well as between the CPU and I/O devices. As a result, data transfer can take place at a time only between CPU and memory, or between CPU and an I/O device. Obviously, this architecture leads to a very slow data transfer. Hence it is used mostly in small computers where performance is not an issue. The main advantage of this architecture is its simple design and ease of programming of I/O device accesses. In this case, I/O devices do not require any special I/O instructions because they use the same memory address space. Hence a WRITE instruction with an address corresponding to an output device will write information to that device. Another advantage of this architecture is that it allows new I/O devices to be easily added to the system, as no new instructions specific to the device are required. Dual Bus Architecture As shown in Figure 7.11, in the dual bus architecture, there are separate set of buses for interconnecting the CPU with memory and with I/O devices. Hence data transfer between the CPU and main memory is carried out by using the buses which interconnect the CPU and main memory, while data transfer between the CPU and I/O devices is carried out by using the buses which interconnect the CPU and I/O devices. In this case, one of the following two mechanisms are used for data transfer between the CPU and an I/O device: 1. Busy Wait. In this method, the CPU sends the I/O request (say a data READ request) to the device via the buses. As already described, the command and the device address associated with the READ request are sent across by using the control bus and the address bus respectively. As soon as the device controller receives the request, it first sets ON a "device busy" indicator so that no further I/O requests for the device will be accepted. It then executes the request and stores the data to be transferred in its local data register. As soon as it cgmpletes the request processing, it sets ON another indicator called "data ready" indicator. After issuing the READ request, the CPU continuously keeps checking the "data ready" indicator of the device and waits until this indicator is set to ON. As soon as the CPU finds this indicator to be ON, it reads the data from the data register of the device controller into its own register. It then sends a data received acknowledgment signal to the device. On receiving this signal, the device controller resets its "device busy" and "data ready" indicators to OFF which indicates that it is now ready to accept new I/O requests.
2. Interrupt. In the busy wait method, we saw that after issuing an I/O request, the CPU remains busy in checking for the completion of the request until the request processing is over. That is, from the time of issuing an I/O request till the completion of the request, the CPU does not do anything else. This is very inefficient use of the CPU because I/O devices are about 1000 times slower than the CPU. The interrupt mechanism has been devised to overcome this problem. In this method, after issuing the READ request to the I/O device, the CPU does not wait for the "data ready" indicator of the device to be set ON. Rather, it suspends the job that made the READ request and takes up another job that is ready for execution and starts executing this job while the device controller is busy processing the READ request. When the device controller completes processing the request and the data to be transferred is ready in its local data register, it sends an interrupt signal to the CPU. When the CPU receives the interrupt signal, it knows that the data it requested is now ready. Hence it suspends the job that it is currently executing, transfers the data from the data register of the device controller to its own register and resumes execution of the job that had made the I/O request. Notice that the interrupt mechanism is more difficult to implement than the busy wait mechanism because it requires a mechanism to suspend a currently executing job and resume it at a later time. How this is done will be discussed in Chapter 14. Moreover, in case of multiple I/O devices, it also requires a mechanism to distinguish among the interrupt signals of different devices and to attend to them based on some priority rules because various devices can initiate interrupt signals independent of each other. DMA Architecture I/O devices are used to input data to or output data from a computer system. Most of the time, this involves either reading data from an input device and storing it in the main memory, or reading data stored in the main memory and outputting it to an output device. That is, data transfer takes, place between the main memory and I/O devices. We saw that in the previous two interconnection architectures, every data transfer to or from an I/O device is carried out by the CPU. Hence the CPU is fully involved in every data transferred between the main memory and I/O devices. This leads to inefficient usage of the CPU. Thus another interconnection architecture, called Direct Memory Access (DMA) architecture, which minimizes the CPU participation in I/O data transfer is often used. As shown in Figure 7.12, in this method an additional hardware called DMA controller [also known as Peripheral Processing Unit (PPU)] is used. All the I/O devices are connected to the DMA controller. A set of buses is used to connect the CPU and memory. Another set of buses is used to directly connect the DMA controller with both the CPU and memory so that data transfer to or from I/O devices can be directly carried out either with the CPU or with the memory. An additional address bus is used between the CPU and the DMA controller. Out of the two address buses, one is used to carry the memory address and the other one is used to carry the device address. Hence in the figure, they have been specifically named as Memory Address Bus (MAB) and Device Address Bus (DAB). The DMA controller has a data register to temporarily store the data being
transferred, a memory address register to keep track of the memory location from/to where data is transferred, a counter to keep track of how many memory words have to be transferred, a "busy" indicator to indicate that it is currently busy and cannot accept any new I/O requests now, and a "data ready" indicator to indicate to the CPU that the requested data is ready in its data register for transfer. The data transfer in this case takes place in the following manner: 1. The CPU sends the I/O request (say a data READ request) to the DMA controller. The device address of the device from which data is to be read, the memory location at which data is to be stored, and the READ command along with the number of memory words to be read are sent across by using the device address bus, the memory address bus, and the control bus respectively. After sending the request, the CPU continues with its normal course of execution since it has no further direct role in this I/O request processing. 2. On receiving the request, the DMA controller sets its "busy" indicator to ON and stores the received memory address and number of memory words to be read in its memory address register and the counter respectively. 3. It then reads the first memory word to be transferred from the specified I/O device into its data register and sets the "data ready" indicator to ON. 4. It now sends an interrupt signal to the CPU. 5. On receiving the interrupt signal, the CPU completes the execution of the instruction that it is currently executing, and then yields the memory address and data buses to the DMA controller for one memory cycle. 6. The DMA controller transfers the data in its data register to the memory location specified" by the contents of its memory address register. It then decrements the value of the counter by one and increments the value stored in its memory address register by one. 7. The DMA controller then reads the next word from the I/O device and stores it in its data register. It then interrupts the CPU, which yields a memory cycle and the memory address and data buses to the DMA controller to enable it to transfer the word from its data register to the appropriate memory location. The DMA controller then decrements the counter value and increments the value stored in its memory address register. This procedure is repeated till the counter value becomes zero. This procedure, where a memory cycle is taken by the DMA controller from the CPU to enable it to transfer data directly to memory is known as cycle stealing. Note that the DMA architecture leads to more efficient CPU utilization because the I/O request is largely processed by the DMA controller without direct participation of the CPU. However, this advantage is gained by providing an extra hardware (the DMA controller) and an additional bus (the device address bus). Points to Remember 1. The CPU is the brain of a computer system. All major calculations and comparisons performed by a computer are carried out inside its CPU. The CPU is also responsible for activating and controlling the operations of other units of the computer system. Hence no
other single component of a computer determines its overall performance as much as the CPU. 2. The two basic components of a CPU are the control unit and the arithmetic logic unit. 3. The control unit of the CPU acts as a central nervous system for the other components of the computer. It manages and coordinates the entire computer system including the input and output units. It obtains instructions from the program stored in the main memory, interprets the instructions, and issues signals that cause other units of the system to execute them. 4. The Arithmetic Logic Unit (ALU) of the CPU is the place where the actual execution of the instructions takes place during the data processing operation. 5. Every CPU has the built-in ability to execute a particular set of machine instructions, called its instruction set. 6. As the instructions are interpreted and executed by the CPU, there is a movement of information between the various units of the computer system. In order to handle this process satisfactorily and to speed up the rate of information transfer, the computer uses a number of special memory units called registers. These registers are used to hold information on a temporary basis and are part of the CPU (not main memory).148 I Foundations of Computing 7. The speed of a processor is directly related to the computer's clock speed, which is the number of pulses, produced each second by the built-in electronic clock. This clock speed is measured in megahertz (or MHz.). 8. Although the processing speed of personal computers is generally measured in MHz, the processing speed of workstations, mini computers, and mainframe systems is often measured in MIPS (Millions of Instructions Per Second) or BIPS (Billions of Instructions Per Second), and that of supercomputers is measured in MFLOPS (Millions of Floating Point Operations Per Second) or GFLOPS (gigaflops which refers to a billion FLOPS). 9. CPUs with large instruction set, variable-length instructions, and a variety of addressing modes are said to employ CISC (Complex Instruction Set Computer) architecture. 10. CPUs with a small instruction set, fixed-length instructions, and reduced references to memory to retrieve operands are said to employ RISC (Reduced Instruction Set Computer) architecture. 11. The Explicitly Parallel Instruction Computing (EPIC) technology is fast emerging as the technology for next-generation processors. The three key features of EPIC technology targeted towards improved performance as compared to today's processors are explicit parallelism, predication and speculation. 12. Every computer has a temporary storage area that is built into the computer hardware and in which instructions and data of a program reside mainly when the program is being executed by the CPU. This storage space is known as primary storage, main memory, or simply memory. 13. Any storage unit of a computer system i,s characterized and evaluated based on the following properties -storage capacity, access time, cost per bit of storage, volatile, random access. 14. A primary storage or main memory of a computer system is made up of several small storage areas called locations or cells. Each of these locations can store a fixed number
of bits called word length of that particular memory. Each location can hold either a data item or an instruction, and its address remains the same regardless of its contents. The addresses normally start at 0 and the highest address equals the number of words that can be stored in the memory minus 1. 15. Data and instructions are moved to and from memory in bunches of word length. Therefore, even if the electronic circuits used are comparable in speed, machines having smaller word-length will be slower in operation than machines having larger word-length. 16. The main memory of some computers is designed to store a fixed number of characters (equal to its word length in bytes) in each numbered address location. Such computers are said to be word-addressable, and they employ a fixed-word-length memory approach. In these computers, storage space is always allocated in multiples of wordlength. In many computers, the main memory is also designed in such a way that each numbered address can only store a single character (A, B, 1, 2, +, -, etc.) Computers designed in this manner are said to be character-addressable and they employ a variableword-length memory approach. 17. The main memory capacity of large computer systems is normally more than small systems. This capacity is defined in terms of the number of bytes a computer system can store. Memory capacity of a computer system is normally stated in terms of kilobytes (KB) which is equal to 1024 (210) bytes of storage, or in terms of megabytes (MB) which is equal to 1,048,576 (220) bytes of storage, or in terms of gigabytes (GB) which is equal to 1,073,741,824 (230) bytes of storage. 18. A computer's main memory is built of volatile RAM chips. RAM chips are of two types - dynamic and static. A dynamic RAM {DRAM), uses an external circuitry to periodically "regenerate" or refresh the storage charge to retain the stored data. 19. A special type of RAM, called read-only memory (ROM) is a non-volatile memory chip in which data are stored permanently and cannot be altered by the programmer. There are two types of read-only memory (ROM) - manufacturer-programmed and userprogrammed. The latter is commonly known as Programmable Read-Only Memory (PROM) because a user can program it. 20. Once information is stored in a ROM chip or a PROM chip it cannot be altered. However, Erasable Programmable Read-Only Memory (EPROM) chips can be reprogrammed to store new information. EPROM chips are of two types - Ultra Voilet EPROM(UVEPROM) and Electrically EPROM(EEPROM). EEPROM is also known as flash memory. 21. The overall performance of a processor can be greatly improved by minimizing the memory-processor speed mismatch. The two commonly used approaches for minimizing the memory-processor speed mismatch are the use of cache memory and memory interleaving. 22. Cache memory is an extremely fast, small memory between CPU and main memory whose access time is closer to the processing speed of the CPU. It acts as a high-speed buffer between CPU and main memory and is used to temporarily store very active data and instructions during processing. 23. The connectivity channel used for interaction between the CPU and memory is known as a bus. Physically, a bus is a set of wires, which carries a group of bits in parallel and has an associated control scheme. The bus width is defined as the number of parallel lines (wires) in the bus. Every computer system has the following
three types of buses - data bus, address bus and control bus. l 24. Just like the buses between the CPU and memory, every computer system also uses buses for interconnecting the CPU (with I/O devices. The same three types of buses (data bus, address bus and control bus) are also used for intercdnnecting the CPU with I/O devices. 25. The interconnection architecture of a computer system defines how exactly its CPU, memory and I/O devices are connected to each other. The accessing mechanism and the flow of data from one unit to another depends on the interconnection architecture. The three commonly used interconnection architectures are unibus architecture, dual bus architecture and DMA (Direct Memory Access) architecture. Questions 1. No other single component of a computer determines its overall performance as much as the CPU. Discuss. 2. List out the main functions of the CPU in a computer system. 3. What are the two main components of the CPU of a computer system? List out the main functions of each of these components. 4. The control unit acts as a central nervous system for the other components of a computer system. Discuss. 5. Describe the role of the decoder of the control unit of a CPU. 6. What is a microprocessor? 7. What is an instruction set of a computer system? Do computers made by different manufacturers generally have the same or different instruction sets? 8. Machine language programs written for one computer will generally not run on another computer with a different CPU. Explain why? 9. What is meant by a family of CPUs? When do two CPUs belong to the same family? 10. When is a computer said to be upward compatible with another computer? How is this feature useful for the users of these computers? 11. What are registers? Name some of the commonly used registers and briefly describe the function of each. 12. What is meant by the length of a register? What are the commonly used registerlengths in modern computer systems? Why is a larger register length more desirable? 13. List out the main steps involved in the execution of an instruction by the CPU of a computer system. 14. What is meant by the clock speed of a computer system? How does it influence the overall speed of a computer system? 15. How is the clock speed of a computer system determined? 16. Differentiate between the terms "clock cycle" and "clock speed". 17. What is a superscalar CPU architecture? How does it lead to a better performance of a computer system? 18. The processing speed of personal computers is generally measured in MHz (Mega Hertz), but that of workstations, mini computers, and mainframe systems is measured in MFLOPS (Millions of Floating Point Operations Per Second). Explain why. 19. Why is the speed of supercomputers generally measured in FLOPS?
20. Name two important design parameters that influence the performance of a CPU. Explain how each parameter influences the speed with which a CPU can execute instructions. 21. List out the key features of CISC processors. 22. List out the key features of RISC processors. 23. Differentiate between CISC and RISC processors. 24. List out the relative advantages and disadvantages of CISC and RISC processors. 25. Explain how RISC architecture leads to the design of faster and less expensive processors as compared to CISC architecture. 26. What is EPIC technology? What are its key features? 27. What is IA-64 architecture? 28. How do EPIC processors have better performance than CISC or RISC processors? 29. What is explicit parallelism? How does it help in improved processor performance? 30. What is predication? How does it help in improved processor performance? 31. Explain the difference between "branch prediction" and "branch predication" techniques. 32. What is load speculation? How does it help in improved processor performance? 33. The performance of EPIC processors comes from both hardware and software. Discuss. 34. Imagine a computer system, which has no primary memory. Explain how would computations be carried out in this system. What would be the major drawback of such a system? Based on this, explain the need for primary memory in a computer system. 35. A fast processor alone is not sufficient for good performance of a computer system. Discuss. 36. List out the key properties used to characterize and evaluate storage units of computer systems. 37. Explain the difference between volatile and non-volatile memory. Give an example of each type of memory. 38. Explain the difference among random access, pseudo-random access, and sequential access storage units. Give an example of each type of storage unit. 39. Differentiate between a bit, a byte, and a word. 40. What is a memory address? Describe with an example. 41. Explain the difference between an address and the contents of an address. 42. Explain the key difference between memory read and write operations. 43. Explain why is it advantageous to have more number of bits per word instead of having more words of shorter length in the memory architecture of a computer system? 44. Distinguish between word-addressable and character-addressable computers. Discuss their relative advantages and disadvantages. 45. How many bytes will be required to store the. word 'MEMORY' in (a) a characteraddressable computer, (b) a word-addressable computer having word-length of 64 bits? 46. Name some of the commonly used units for stating memory capacity. Also state the total number of characters each unit is roughly equivalent to. 47. A computer has 512 MB of memory. How many characters can be stored in its memory at a time? 48. Is it possible to enhance the existing memory capacity of a computer system? If no, explain why. If yes, explain how.
49. Differentiate between static and dynamic RAMs. Which one is commonly used for the main memory of most computers? 50. What is a ROM? Why is it so called? Give few typical uses of ROM. 51. What is a microprogram? Give an example of a microprogram that is a good candidate for storage in the ROM of a computer system. 52. Differentiate between manufacturer-programmed and user-programmed ROMs. 53. Differentiate between PROM and EPROM. 54. Differentiate between UVEPROM and EEPROM. In which case is it easier to alter stored information and why? 55. What is a flash memory? Why is it so called? Give its few typical uses. 56. Differentiate among RAM, ROM, PROM and EPROM. 57. Write short notes on: (a) CISC architecture (f) EPROM (b) RISC architecture (g) Memory capacity (c) RAM (h) Instruction set (d) Processor speed (i) EPIC technology (e) ROM 58. The rate at which data can be fetched from memory is about ten times slower than the rate at which CPU can process data. Name two methods that may be used to minimize this speed mismatch. Give the details of one of the two methods and explain how it helps in minimizing the speed mismatch. 59. What is a cache memory? How is it different from a primary memory? 60. Explain how a cache memory helps in improving the overall processing speed of a computer system. 61. The effective use of cache memory requires that the data and instructions needed for processing by the CPU are made available beforehand in the cache. Explain how is this ensured in the system. 62. What is the principle of locality of reference? How is it related to the speed of execution of a program? In what manner do branch statements affect the locality of reference? 63. Many computer systems use separate data and instruction caches. Explain why. 64. Explain the concept of multi-level cache design in a computer system. What are its advantages and disadvantages as compared to a single-level cache design? 65. What are the key design issues involved in the use of cache memory in a computer system? Describe the commonly used mechanisms to handle each of these issues. 66. What is cache size? How is it normally decided? 67. What is meant by block size in the context of cache design? How is it normally decided? 68. What is hit ratio? How can a programmer help in improving the hit ratio of a computer system? 69. What is meant by replacement policy in the context of cache design? Name a commonly used cache replacement policy. 70. Explain how the least recently used (LRU) replacement policy leads to effective cache usage. 71. What is meant by write policy in the context of cache design? Describe two write policies and list their relative advantages and disadvantages.
72. What is memory interleaving? Explain how it helps in improving the overall processing speed of a computer system. 73. Explain how instructions are distributed in different modules of an interleaved memory system. 74. Explain how cache memory and interleaved memory mechanisms can be used together to improve the overall processing speed of a computer system. 75. What is a bus? Why is it required in a computer system? 76. What is meant by bus width? How does it affect the overall speed of a computer system? 77. What are memory buses? How many types of memory buses are there? Describe the role of each type. 78. A wider bus enables faster exchange of data. Discuss. 79. Why is the width of the memory address bus an important parameter for a computer system? 80. How are the width of the memory address bus and the number of memory locations of a computer system co-related? 81. What are I/O buses? How many types of I/O buses are there? Describe the role of each type. 82. List out the similarities and differences among the ISA, EISA and MCA buses. 83. What is an I/O device controller? Explain its role in I/O data transfer. 84. What is meant by the interconnection architecture of a computer system? List out some of the commonly used interconnection architectures. 85. Describe how data transfer takes place in case of unibus interconnection architecture. List out some of the advantages and limitations of this architecture. 86. Describe the busy wait mechanism of data transfer between the CPU and an I/O device in case of dual bus architecture. List out the advantages and limitations of this mechanism. 87. Describe the interrupt mechanism of data transfer between the CPU and an I/O device in case of dual bus architecture. List out the advantages and limitations of this mechanism. 88. Draw the schematic diagram of the DMA interconnection architecture. Describe how data transfer takes place in this case. Explain how this architecture minimizes the CPU participation in I/O data transfer. 89. List out the advantages and limitations of DMA data transfer mechanism.154 I Foundations of Computing 90. What is cycle stealing? 91. Write short notes on: (a) Central Processing Unit (CPU) (b) Arithmetic Logic Unit (ALU) (c) Registers (d) Storage evaluation criteria (e) I/O buses (f) Static and Dynamic RAM (g) Cache memory (h) Locality of reference (i) Cache replacement policy
(j) Interleaved memory (k) Memory buses (1) Volatile and Non-volatile storage (m) Interconnection architectures (n) DMA data transfer (o) Fixed and Variable word-length memory 92. Write the full form of the following abbreviations: (a) KB (n) PROM (b) RAM (o) EPROM (c) ROM (p) MAR (d) MBR (q) GFLOPS (e) MB (r) FLOPS (0 GB (s) SIMM (g) CISC (t) LRU (h) RISC (u) ISA bus (i) SRAM (v) EISA bus (j) DRAM (w) MCA bus (k) UVEPROM (x) PPU (1) EEPROM (y) DMA (m) MFLOPS (z) EPIC
Chapter 8 Secondary Storage Devices It is desirable that the operating speed of the primary storage of a computer system be as fast as possible because most of the data transfer to and from the processing unit is via the main memory. For this reason, storage devices with very fast access times, such as semiconductors, are generally chosen for the design of primary storage. These high-speed storage devices are expensive and hence the cost per bit of storage is also high for a primary storage. Moreover, these storage devices are volatile. That is, the data stored in them are lost when the power is turned off or interrupted. Hence the primary storage of a computer system has the following limitations: 1. Limited Capacity. The storage capacity of the primary storage is limited. Often, it is necessary to store many millions, sometimes billions, of bytes of data. Unfortunately, the storage capacity of the primary storage of today's computers is not sufficient to store the large volume of data handled by most data processing centres. 2. Volatile. The primary storage is volatile and the data stored in it is lost when the electric power is turned off or interrupted. However computer systems need to store data on a permanent basis for several days, several months, or even several years. As a result, additional memory, called auxiliary memory or secondary storage, is used with most computer systems. This section of the computer's memory is non-volatile and has low cost per bit stored, but it generally has an operating speed far slower than that of the primary storage. This section of the memory is used to store large volume of data on a permanent basis which can be partially transferred to the primary storage as and when required for processing. Data is stored in secondary storage in the same binary codes as in main storage and is made available to main storage as needed. Over the years, a wide range of devices and media have been developed for use as secondary storage in computer systems. Many of them, such as punched paper tape and punched cards, have become obsolete. As shown in Figure 8.1, the popular ones used in today’s computer systems are magnetic tape, floppy disk, zip disk, disk pack, Winchester disk, CD-ROM, WORM disk and magneto-optical disk. In this chapter you will learn about the terminologies, principles of operation, uses, and trade-offs of many different types of secondary storage devices. SEQUENTIAL AND DIRECT-ACCESS DEVICES Several different devices can be used as a secondary storage device, but the one selected for a particular application mainly depends upon how the stored information needs to be accessed. Basically, there are two methods of accessing information - sequential or serial access and direct or random access. A sequential-access storage device is one in which the arrival at the location desired may be preceded by sequencing through other locations, so that access time varies according to location. In other words, information on a
sequential-access device can only be retrieved in the same sequence in which it is stored. Sequential processing is quite suitable for such applications like preparation of monthly pay slips, or monthly electricity bills, etc., where most, if not all, of the data records need to be processed one after another. In these applications, data records for each and every employee or customer needs to be processed at scheduled intervals (in this case monthly). However, while working with a sequential-access device, if an address is required out of order, it can only be reached by searching through all those addresses which are stored before it. For instance, data stored at the last few locations cannot be accessed until all preceding locations in the sequence have been traversed. This is analogous to a music tape cassette. If 10 songs are recorded on a music tape and if you want to listen to (or access) the 8th song, you must listen to the 7 songs that come before your favorite song before you can hear it. Although your player may "fast forward" the tape quickly past the first 7 songs, the first 7 songs are still accessed although not fully played. In the same way, to access the 28th data record stored on a sequential-access device, the computer must first access (although not process) the first 27 data records. Magnetic tape is a typical example of a sequential-access storage device. In many applications, we need to access information in a more direct manner than sequential-access devices allow. For example, in a computerized bank, at any instance it is required to determine the exact balance in the savings account of a particular customer. Similarly, in a computerized airline ticket booking system, immediate access may be required to reservation system records to find out if seats are currently available on a particular flight. In such applications, if we use a sequential-access device for data storage, the time taken to access the desired information may be enormous which will cause frustration to the customer. Secondary storage devices exist which permit access to individual information in a more direct or immediate manner. These direct-access devices are also called random-access devices because the information is literally available at random, i.e., it is available in any order. Thus a random-access storage device is one in which any location in the device may be selected at random, access to the information stored is direct, and approximately equal access time is required for each location. This is analogous to a music CD. If 10 songs are recorded on a music CD and if you want to listen to (or access) the 6th song, you need not listen to or fast forward the 5 songs that come before your favorite song. All you have to do is select track number 6 on the player and the player directly moves the pickup arm across the CD to the groove where the 6 th song begins and starts playing. Magnetic, optical and magneto-optical disks of various types are typical examples of direct-access storage devices. MAGNETIC TAPE Magnetic tape is one of the oldest forms of secondary storage device. It is one of the most popular storage medium for large data that are sequentially accessed and processed or that are not used very often and need to be stored off-line (not directly accessible to the computer system). It is also used as a backup storage (additional copy of data kept away from the computer system) for data stored on on-line storage devices such as a hard disk. The magnetic tape medium is a plastic ribbon usually XA inch or % inch wide and 50 to 2400 feet long. It is coated with a magnetizable recording material such as iron-oxide or
chromium dioxide. Data are recorded on the tape in the form of tiny invisible magnetized and non-magnetized spots (representing Is and 0s) on the coated surface of the tape. The tape ribbon is itself stored in reels or a small cartridge or cassette. Basic Principles of Operation Storage Organization The tape of a magnetic tape storage is divided into vertical columns called frames and horizontal rows called channels or tracks. As shown in Figure 8.2, older tapes had 7 tracks and they used the 6-bit BCD code format for data recording. As in BCD code, the letter A is represented on this tape by the code 110001. The first six tracks are used for recording the 6 bits of the BCD code. The seventh track is used for recording the parity bit. A parity bit or check bit is used to detect errors that may occur due to the loss of a bit from a string of 6 or 8 bits during data input or output operations. If the basic code for a character requires an odd number of 1 bits (such as the characters 1, 2, or A in BCD) an additional 1 bit is added to the check bit location so that there will always be an even number of 1 bits. This is an example of even-parity. Similarly, in odd-parity, the check bit is used to always produce an odd number of 1 bits. That is, the check bit will be 1 if the total number of 1 bits for representing a particular character is even and it will be 0 otherwise. The tape shown in Figure 8.2 uses the parity bit for an even parity. Most of the modern magnetic tapes have 9 tracks and they use the 8-bit EBCDIC code format for data recording. As shown in Figure 8.3, in this case the fourth track is used for parity bit, which produces an odd parity in this example. Notice that the letter A is represented here by the code 11000001 as in the 8-bit EBCDIC code format. A magnetic tape is a continuous medium in which data are recorded serially. There is no addressing. In this situation, how can different pieces of data (known as records) be identified on a tape? The answer is that records may be separated by blank spaces on the tape called inter-record gaps (IRG). These gaps are automatically created when data are written on the tape. Each gap is of the order of 0.5 inches. When data are read from a moving tape, the tape movement stops when a gap is reached. The tape remains motionless until the record is processed and then moves again to read the next record. This procedure is repeated until all the records in a file are processed. Records stored on tapes can be of varying lengths. If a tape contains a large number of very short records with an IRG after each record, then more than half of the tape could be unused (occupied with IRGs), and the tape I/O operation would also be very slow due to very frequent interruption in tape movement during I/O operation. To overcome these problems of inefficient usage of tape storage space and inefficient I/O operation, records are usually grouped in blocks of two or more, separated by an inter-block gap {IBG). The process of grouping two or more records together to form a block of data is known as blocking and the number of records put together in each block is called blocking factor.
The blocking factor generally depends on the record length. The rule of thumb for deciding blocking factor is that the block should be at least ten times as long as the IBG to reduce wastage of tape storage space. Blocking also helps in faster tape I/O operation because it permits additional records to be transmitted with each "read" or "write" operation. For each tape "read" operation, all data records between adjacent IBGs are read at a time into the main memory of the computer system for processing. The next "read" operation transmits the next block of records to the main memory. For a "write" operation on a tape, an entire block of records are transmitted from the main memory to the tape drive. Then a block of data and an IBG are written on the tape. Figure 8.4 illustrates the concept of blocking. Figure 8.4 (a) shows an unblocked tape (one which does not use blocking), Figure 8.4 (b) shows a tape which uses a blocking factor of two, and Figure 8.4 (c) shows a tape which uses a blocking factor of three. Notice how the tape usage increases as the blocking factor increases. However, the blocking factor should not be too large otherwise in case of data loss due to some parts of the tape becoming bad, all the records in a block will become irretrievable. A group of records form a block, and a set of blocks constitute a file. A single tape is normally used to store several files. In this case we face the problem of uniquely identifying the files stored on a tape and separating one file from the other when they are sequentially stored one after another on the tape. To take care of this problem, the computer adds a file header label and a file trailer label respectively at the beginning and at the end of a particular tape file. The file header label identifies the beginning of a file and contains the file's attributes such as the name of the file (file identifier), the date it was created, etc. The file trailer label signifies the end of a file and contains the total number of records in the file. The data organization on a magnetic tape along with file header and file trailer labels is shown in Figure 8.5. A tape is wound on a spool and during tape I/O operation it moves from a supply spool to a take-up spool. To allow for threading on the take-up and supply spools, the first and the last several feet of tape are unused (not used for any type of information recording). After the unused length towards the beginning of tape, a reflective metal foil known as the beginning of tape (BOT) marker is used to indicate to the equipment the beginning of usable tape. Similarly, before the unused length towards the end of tape, another reflective metal foil known as the end of tape (EOT) marker is used to indicate to the equipment the end of usable tape. These two markers are placed on opposite edges of the tape for machine identification purposes. Between the BOT marker and the first file header label is a tape header label, which contains the tape's attributes such as the tape identifier, the number of files it contains, the date it was last used, and other control information that helps to prevent an important tape from accidentally being erased. The data organization on a magnetic tape along with BOT and EOT markers and tape header label is shown in Figure 8.6.
Storage Capacity The storage capacity of a tape depends on the following: 1. Data recording density, and 2. Tape length It is a multiple of the above two parameters. That is, Storage capacity of a tape - Data recording density * Length Data recording density refers to the amount of data that can be stored on a given length of tape. It is measured in bytes per inch (bpi), or the number of bytes (characters) that can be stored per linear inch of tape. Tape density varies from 800 bpi in older systems to 77,000 bpi in some of the most modern systems. Hence if a 2400 feet tape has a data recording density of 800 bpi, then its storage capacity will be 2400 * 12 inches x 800 bpi = 23 x 106 bytes = 23 Giga bytes. Note that this is the total storage capacity of the tape. Its actual storage capacity (storage available for storing user data) is much less due to the use of IBGs, file header labels, file trailer labels, BOT and EOT markers. In fact, the actual storage capacity of a tape may be anywhere from 35% to 70% of its total storage capacity, depending on the storage organization. Data Transfer Rate Data transfer rate refers to the number of characters per second that can be transmitted to the primary storage from the tape. It is measured in bytes per second (bps). Its value depends on the data recording density and the speed with which the tape travels under the read/write head. Tape speed is typically of the order of 100 inches per second. Therefore, a tape having data recording density of 77,000 bpi and its drive having a tape speed of 100 inches per second, will have a data transfer rate of 77,000 x 100 = 77,00,000 bytes or 7.7 MB per second. Tape Drive A magnetic tape drive is used for storage and retrieval of data stored on magnetic tape medium. The tape drive is different for tape reels, tape cartridges, and tape cassettes. However, all of them work on a similar mechanism like the audio tape recorders or video cassette recorders (VCR) found in our homes. That is, the tape drive has read/write heads and as the tape ribbon passes under the read/write heads, the data can be either read and transmitted to primary storage, or transmitted from primary storage and written to the tape by giving suitable commands to the tape drive. Instead of play and record, read and write commands are used with the tape drive. Just as in the case of an audio tape recorder or video cassette recorder, a magnetic tape reel or cartridge or cassette has to be first loaded to a tape drive for processing. Once loaded to a tape drive, the magnetic tape is said to be on-line; that is, it can now be used for storage or retrieval of data by the computer system. When processing is complete, the tape is removed from the tape drive for off-line storage; that is, it is stored away from the computer system and data on it are not accessible to the computer system until it is loaded again on the tape drive for processing.
Like audio or video tape, the magnetic tape, used in computer systems can also be erased and reused indefinitely. Old data on a tape are automatically erased as new data are recorded in the same area. Tape Controller A magnetic tape has to be mounted on a tape drive before it can be used for reading or writing of information. A tape drive is connected to and controlled by a tape controller which interprets the commands for operating the tape drive. A typical set of commands supported by a tape controller are: Read reads one block of data Write writes one block of data Write tape header label used to update the contents of tape header label Erase tape erases the data recorded on a tape Back space one block rewinds the tape to the beginning of previous block Forward space one block forwards the tape to the beginning of next block Back space one file rewinds the tape to the beginning of previous file Forward space one file forwards the tape to the beginning of next file Rewind fully rewinds the tape Unload releases the tape drive's grip so that the tape spool can be unmounted from the tape drive Types of Magnetic Tapes The data recording density and data transfer rate of magnetic tapes depend to a large extent on the data organization and principles of operation of various types of magnetic tapes and their tape drives. The commonly used ones are: 1. ½ - inch tape reel 2. ½ -inch tape cartridge 3. ¼ inch streamer tape 4. 4-mm digital audio tape (DAT) They are described below. Half-inch Tape Reel A magnetic tape reel uses 14 inch tape ribbon stored on a tape reel (see Figure 8.7). The magnetic tape drive of a tape reel is shown in Figure 8.8. The tape on a reel moves through a tape drive in much the same way as a video tape moves through a movie projector. During processing, the tape moves from a supply reel to a take-up reel via two vacuum channels and through a read/write head assembly. The read/write head assembly either reads data, or writes data on the tape. It is a single unit having one read/write head for each tape track. Therefore, the read/write head assembly of a 9-track tape drive will have a set of nine heads. Each head operates independently and stores information on nine parallel tracks, parallel to the edge of the tape. This method of storing data in adjacent bit configurations is known as, parallel representation. In parallel representation, data are read or written a byte at a time.
The two vacuum channels of the tape drive are designed to take up slack tape, acting as buffers to prevent the tapes from snapping or stretching when starting from a stationary position or slowing down from full speed. Several methods are used to prevent tape damage from sudden bursts of speed and different tape drives may use different mechanisms for this purpose. Magnetic tape reels are commonly used as archival storage for off-line storage of data and for exchange of data and programs between organizations. They are mainly used in organizations having large computing systems such as mainframe systems because the tape controller and tape drives used for magnetic tape reels are bulky and expensive. Many medium-sized business organizations have vaults containing several thousands of these tape reels. For example, a large insurance company may have a storage area containing over 50,000 magnetic tape reels. However, the magnetic tape reels are fast getting replaced by tape cartridge, streamer tape, and digital audio tape because these new magnetic tape storage devices are more compact, cheaper, and easier to handle and use. Half-inch Tape Cartridge Magnetic tape reels are suitable for use only with large and medium size computers. The same magnetic tape is used in smaller computer systems in the form of tape cartridges. As shown in Figure 8.9, these tape cartridges look very similar to the video cassettes used in the familiar home VCRs (video cassette recorders). The tape drives of these tape cartridges are also very similar in look and mechanical operation to the familiar home VCRs. A tape cartridge can be pushed inside or ejected from the tape drive just as a video tape cassette is pushed inside or ejected from a VCR unit. These tape cartridges provide a more convenient way to package tape, and greatly simplify the mounting of tape which is normally a problem with magnetic tape reels that have to be manually mounted on the tape drive. Moreover, tape cartridges also provide protection against dirt and contamination, since the tape is sealed inside the cartridge. Aside from the above mentioned differences in the mechanical shape and size of halfinch tape reels and half-inch tape cartridges, and the mechanical operation of their tape drives, there are two main differences in their data organization and data recording/reading: 1. The half-inch tape cartridge has 36 tracks, as opposed to 9 tracks for most half-inch tape reels. Like tape reels, tape cartridges also store data using parallel representation. Hence, 4 bytes of data are stacked across the width of the tape in case of tape cartridges, as opposed to a single byte of data in case of tape reels. This enables more bytes of data to be stored on the same length of tape. 2. Tape cartridges record/read on the top half of the tape in one direction and on the bottom half in the other direction. That is, out of 4 bytes (36 tracks) of data stacked
across the width of the tape, 2 bytes (18 tracks) of data are read in each direction. Recall that in case of tape reels, data are recorded/read only in one direction. Other principles of data organization on half-inch tape cartridges are same as that for half-inch tape reels. That is, in case of half-inch tape cartridges also the concepts of blocking of records, IBG, file header label, file trailer label, tape header label, BOT marker, EOT marker, etc., are used for their respective purposes. Quarter-inch Streamer Tape In addition to the storage of data of those applications which process data sequentially, one of the primary uses of magnetic tapes is to serve as a backup storage medium for data stored on on-line devices such as a disk. During backup or recovery process, backup tapes are normally processed continuously from beginning to end because there is seldom a need for selective access of records. Hence there is no need to start and stop the backup tape after every few records. Based on these observations, researchers came out with a special type of quarter-inch streamer tapes that eliminate the need for the start/stop operation of traditional tape drives for half-inch tape reels or tape cartridges. These tapes are J4 inch wide and are sealed in a cartridge. Their tape drives are very similar in look and mechanical operation to the tape drives of half-inch tape cartridges. Unlike the half-inch tapes, which use parallel representation for data recording, these tapes use serial representation for data recording. That is, the data bits are aligned in a row, one after another in tracks. A streamer tape can have from 4 to 30 tracks, depending on the tape drive. Depending on the tape drive, the read/write head reads or writes data on one, two, or four tracks at a time. Figure 8.10 shows a streamer tape having 8 tracks on which data are recorded two tracks at a time. That is, data are first written on the top two tacks for the entire length of the tape or until the data are exhausted. If the end of the tape is reached while data recording, the tape is rewound, the read/write head is positioned on the next two tracks, and data recording continues in a similar manner. As there is no start/stop mechanism, streamer tape drives can read/write data more efficiently than the traditional tape drives that stop and restart the tape movement at each IBG. Moreover, streamer tape drives also make more efficient utilization of tape storage area. Typically, streamer tapes use 97% of the tape for data storage, whereas traditional start/stop tapes use only 35% to 70%, depending on the blocking factor. Typical capacities of streamer tape drives used in personal computers varies from 50 MB to 500 MB. Streamer tapes are so called because they read/write data serially as streams of bits. The data format used in these tapes has been standardized by the industry and is known as the QIC standard. This helps in easy use of these tapes for exchanging data between different computers. That is, data recorded on a streamer tape by one computer can be read by another computer if both of them use the common QIC standard data format. QIC stands for Quarter Inch Cartridge.
4mm Digital Audio Tape (DAT) This is the most recent type of magnetic tape storage device which provides a very high data density per inch of tape. It uses a tape ribbon of 4mm width enclosed in a cartridge. The tape length is either 60 meters or 90 meters. Very high data recording density is obtained by using a tape drive that uses helical scan technique for data recording. As shown in Figure 8.11, the helical scan has two read heads and two write heads built into a small wheel (or cylinder) that spins near the tape at about 2000 rpm. At the same time, the tape also moves past the wheel at a relatively slow speed of about ⅓. inch per second. The write heads on the spinning wheel each write data with opposite magnetic polarities on overlapping areas of the tape. Each read head reads only one polarity or the other. The result is a very high data density per inch of tape. The helical scan records at an angle to the tape. DAT drives use a data recording format called Digital Data Storage (DDS) which provides three levels of error correcting code to ensure excellent data integrity. Typical capacity of DAT cartridges varies from 4 GB to 14 GB. DAT drives have a data transfer speed of about 366 KB/sec. Advantages and Limitations of Magnetic Tapes Advantages As a secondary storage device, magnetic tapes have the following advantages: 1. Their storage capacity is virtually unlimited because as many tapes as required can be used for storing very large data sets. 2. With the low cost of tape reels and cartridges and high data recording densities, the cost per bit of storage is very low for magnetic tapes. An additional cost benefit is that tapes can be erased and reused many times. 3. Since the tape reels and cartridges are compact and light in weight, they are easy to handle and store. Very large amount of data can be stored in a small storage space. 4. Due to their compact size and light weight, magnetic tape reels and cartridges are also easily portable from one place to another. They are often used for transferring data and programs from one computer to another, which are not linked together. Limitations As a secondary storage device, magnetic tapes have the following limitations: 1. Due to their sequential access nature, they are not suitable for storage of those data that frequently require to be accessed randomly. 2. They must be stored in a dust-free environment because specks of dust can cause tape-reading errors. 3. They must also be stored in-an environment with properly controlled temperature and humidity levels, otherwise the tape ribbon may get twisted due to warping, resulting in loss of stored data. 4. They should be properly labeled so that some useful data stored on a particular tape is not erased by mistake.
Uses of Magnetic Tapes Magnetic tapes are typically used for one or more of the following purposes: 1. For applications that are based on sequential data processing. For example as shown in Figure 8.12, the contents of an inventory master file (which is the permanent source of inventory data) may be periodically updated using a transaction file (which contains data reflecting inventory activities during the period) to create a new inventory master file which is used in the next processing cycle. Magnetic tapes are the most suitable storage for such applications. 2. Backing up of data stored on an on-line storage device such as a disk for its off-line storage so that if by accident the data on the disk gets corrupted or gets lost, it can be retrieved from the backup tape and stored back on the disk. . 3. Archiving of data which are not used frequently, but which may be used once in a while. 4. Transferring of data and programs from one computer to another, which are not linked together. 5. Distribution of software by vendors. Originally sold software or software updates are often distributed by vendors on magnetic tapes. MAGNETIC DISK Magnetic disks are the most popular medium for direct-access secondary storage. Because of their random access capability, they are the most popular on-line secondary storage devices. A magnetic disk is a thin, circular plate/platter made of metal or plastic that is usually coated on both sides with a magnetizable recording material sych as iron-oxide. Data are recorded on the disk in the form of tiny invisible magnetized and non-magnetized spots (representing Is and Os) on the coated surfaces of the disk. A standard binary code, usually 8-bit EBCDIC, is used for recording data. The disk itself is stored in a specially designed protective envelope or cartridge, or several of them may be stacked together in a sealed, contamination-free container. Like magnetic tapes, magnetic disks can also be erased and reused indefinitely. Old data on a disk are automatically erased as new data are recorded in the same area. However, the information stored can be read many times without affecting the stored data. Hence the reading operation is non-destructive of stored data, and the data stored on a magnetic disk remains indefinitely until they are erased or over written at a future time. Basic Principles of Operation Storage Organization For data recording, the surface of a disk is divided into a number of invisible concentric circles called tracks. As shown in Figure 8.13, the tracks are numbered consecutively from outermost to innermost starting from zero. The number of tracks varies greatly
between disks, from as few as 40 on some small, low-capacity disks to several thousand on large, high-capacity disks. Each track is further subdivided into sectors. For this, in addition to the concentric circles, the disk surface is also divided into invisible pie-shaped segments (see Figure 8.14). Thus if there are eight such pie-shaped segments, each track will get divided into eight parts, and each of these eight portions of a track is called a sector. A sector typically contains 512 bytes. A sector is the smallest unit with which any disk can work. That is, disk drives are designed to access (read/write) only whole sectors at a time. Even if the computer needs to change just one byte out of the 512 bytes stored on a sector, it rewrites the entire sector. Notice from Figure 8.14 that the sectors on the outer tracks are longer than those on the inner ones. Since each sector records the same number of characters (usually 512 characters), the length of the sectors on the innermost track determines the storage capacity for the sectors on all other tracks of the disk, This means that the storage space remains underutilized on the sectors in the outer tracks of the disk. This is a disadvantage of dividing disks into pie-shaped segments. To avoid this problem of underutilization of storage space, some systems such as Apple and Commodore, partition the disk in such a manner that all sectors occupy approximately the same amount of space. One such type of partitioning of disk is shown in Figure 8.15. In this method, the tracks are first partitioned into groups, and the tracks in different groups have different number of sectors per track. For example, in Figure 8.15 the tracks are partitioned into three groups. The tracks belonging to the innermost group have 8 sectors per track, the tracks belonging to the next group have 9 sectors per track and the tracks belonging to the outermost group have 10 sectors per track. Using this new method of disk partitioning, Apple was able to increase the storage capacity of its 3 Vi-inch floppy disk by about 40%. When people refer to the number of sectors a disk has, the unit they use is sectors per track - not just sectors. Thus if a disk has 200 tracks and 8 sectors per track, it has 1600 (200 x 8) sectors - not 8 sectors. Each sector of a disk is assigned a unique number. Before a disk drive can access a piece of data (a record) stored on a disk, it must specify the record's disk address. The disk address is comprised of the sector number, track number, and surface number (when double-sided disks are used). That is, the disk address represents the physical location of the record on the disk.170 We saw that disk drives are designed to access only whole sectors at a time (a sector being the smallest unit of data access from a disk). Since a file usually is not of a size that is an even multiple of sector size (usually 512 bytes), some part of the last sector allocated to the file may be unused. Thus, on an average half of the sector size is wasted for each file stored on the disk. Hence it is recommended that the sector size for a disk should not be too large. The sector size should not even be too small, otherwise the size
of the mapping table used to map disk addresses to data records will become too large. The DOS operating system goes one step further of combining two or more sectors to form a cluster. In this case, the smallest unit of data access from a disk becomes a cluster -not a sector. That is, read/write operations read/write a whole cluster at a time. Cluster sizes vary, depending on the size and type of the disk, but they can range from 2 to 64 sectors per cluster. Notice that a cluster-based disk organization may lead to more wastage of disk space as compared to sector-based organization, because in case of cluster-based systems, on an average, half of the cluster size is wasted for each file stored on the disk. However, cluster-based organization, leads to less management overhead for mapping of data records to physical disk locations. Often multiple disks are stacked and used together to create large capacity disk storage systems. In this case, a set of magnetic disks is fixed to a central shaft one below the other to form a disk pack (see Figure 8.16). The disk pack is sealed and mounted on a disk drive, which consists of a motor to rotate the disk pack about its axis. The disk drive also has an access arm assembly that has separate read/write heads for each surface of the disk pack on which data can be recorded. Normally, the upper surface of the top disk and the lower surface of the bottom disk are not used in a disk pack because these surfaces may get easily scratched. The access arm assembly is designed in such a way that all the access arms (on which the read/write heads are fixed) for all the disk surfaces move together. Thus, if the read/write head that serves the 0th recording surface is positioned over the 5th track, each of the heads on the arms that serve other recording surfaces is similarly positioned over the 5th track. For faster access of data from disk packs, a concept called cylinder is used for data organization on disk packs. As shown in Figure 8.16, a set of corresponding tracks on all the recording surfaces of a disk pack together form a cylinder. For example, the 5th track of all the recording surfaces together comprise the 5th cylinder of the disk pack. Thus if there are 200 tracks on a single disk surface, there are 200 cylinders in the disk pack. Obviously, in this case the disk address of a data record is comprised of sector number, cylinder number, and surface number (track number is not required because track number and cylinder number are same). Faster access of data is achieved by cylinder-based organization by avoiding movement of access arms when large number of related records are to be processed in sequence. The related records of a file can be stored on the same cylinder of the multiple disks of a disk pack so that in one revolution of the disk pack, all records stored on say track 5 of surface 0 are read. In the next revolution, the subsequent records stored on track 5 of surface 1 are read, in the revolution after that, the subsequent records stored on track 5 of surface 2 are read, and so on. This procedure can continue down the cylinder without any delays caused by the movement of access arms. Storage Capacity As you might have already guessed, the storage capacity of a disk system depends on the following: 1. Number of recording surfaces, 2. Number of tracks per surface, 3. Number of sectors per track, and
4. Number of bytes per sector It is a multiple of the above four parameters. That is, Storage capacity of a disk system = Number of recording surfaces x Number of tracks per surface x Number of sectors per track x Number of bytes per sector For example, suppose a disk pack has 10 disk plates each having 2655 tracks. Suppose there are 125 sectors per track, and each sector can store 512 bytes. Since the disk pack has 10 disk plates, so it will have 18 recording surfaces (excluding the upper surface of the topmost disk and the lower surface of the bottommost disk). Hence the capacity of this disk pack will be = 18 x 2655 x 125 x 512 = 3,05,85,60,000 bytes = 3 x 10 9 bytes (approximately) = 3 GB (3 Giga Bytes). Essentially one character can be stored per byte. So a disk system with 3 GB capacity can store 3 billion characters of information. Notice that the more recording surfaces a particular disk system has, the greater its storage capacity will be. For example, in case of a disk system having only one disk (i.e., a floppy disk), the storage capacity of a double-sided disk plate will be about twice that of a single-sided one. And the storage capacity of a disk system having multiple disk plates (i.e., a disk pack) will be many times more than a single-disk system. As the storage capacity of a disk system is also directly proportional to the number of tracks per surface, disks having larger diameter can have more tracks and larger storage capacity. However, disk size cannot be made too large because it will affect the overall size of the disk unit. And the constant effort of the designers of disk systems is to provide more storage capacity in as compact form as possible. Hence instead of increasing the disk size for larger storage capacity, the designers prefer to increase the storage capacity by increasing th& tracks per inch of surface and bits per inch of track. That is, a constant goal of the designers of disk systems is to increase the data density of each square inch of disk surface. Access Mechanism Data are recorded on the tracks of a spinning disk surface and read from the surface by one or more read/write heads. As shown in Figure 8.17, the read/write heads are mounted on an access arms assembly. Most disk drives have a single read/write head for each disk surface. But some faster disk systems use multiple heads on each access arm to service a number of adjacent tracks simultaneously. The access arms assembly can be moved in and out in the direction shown in the figure so that the read/write heads can be moved horizontally across the surfaces of the disks. In this manner, the read/write heads can be positioned on any track on which data are to be recorded or from which data are to be read. In case of a disk pack, each usable surface has its own read/write head and all the heads move together. Hence information stored on the tracks, which constitute a cylindrical shape through the disk pack, is accessed simultaneously. Recall the cylindrical storage arrangement of information in a disk pack.
The read/write heads are of flying type. That is, they do not have direct contact with the disk surfaces. This prevents wear on the surface of the disk. A separation of about 0.00002 inch is maintained between a read/write head and its corresponding disk surface. In fact, the read/write heads fly so close to the disk surface that if a dust particle (typically of 0.0015 inch size), smoke particle (typically of 0.00025 inch size), finger print (typically of 0.00062 inch size), or a human hair (typically of 0.003 inch size) were placed on the disk surface, it would bridge the gap between the read/write head and the disk surface, causing the head to crash. A head crash, in which the head touches the disk, destroys the data stored in the area of the crash and can destroy a read/write head as well. Access Time Disk access time is the interval between the instant a computer makes a request for transfer of data from a disk system to the primary storage and the instant this operation is completed. To access information stored on a disk, the disk address of the desired data has to be specified. We saw that the disk address is specified in terms of the surface number, the track/cylinder number, and the sector number. Information is always written from the beginning of a sector and can be read only from the track beginning. Hence disk access time depends on the following three parameters: . 1. Seek Time. As soon as a read/write command is received by the disk unit, the read/write heads are first positioned on to the specified track (cylinder) number by moving the access arms assembly in the proper direction. The time required to position the read/write head over the desired track is called the seek time. The seek time varies depending on the position of the access arms assembly when a read/write command is received. If the access arms assembly is positioned on the outer most track and the track to be reached is the inner most one then the seek time will be maximum, and it will be zero if the access arms assembly already happens to be on the desired track. The average seek time is thus specified for most systems which is of the order of 10 to 100 milliseconds. We also saw that some disk systems have multiple read/write heads on each access arm. This is basically for reducing the seek time. For example, a disk system may have two sets of read/write heads for each surface, one for reading/writing on the inside tracks and another for the outside tracks. This will reduce the average seek time by half of that for a disk system which has a single read/write head per surface, because each read/write head needs to cover and move across only half of the total number of tracks. 2. Latency. Once the heads are positioned on the desired track, the head on the specified surface is activated. Since the disk is continuously rotating, this head should wait for the desired data (specified sector) to come under it. This rotational waiting time, i.e., the time required to spin the desired sector under the head is called the latency. The latency, also known as the rotational delay time, is also a variable and depends on the distance of the desired sector from the initial position of the head on the specified track. It also depends on the rotational speed of the disk which may be anywhere from 300 rpm (rotations per minute) to 7200 rpm. An average latency time is thus normally specified
which is of the order of 5 to 80 milliseconds. Note that the average latency of a disk system is equal to half the time taken by the disk to rotate once. Hence the average latency of a disk system whose rotational speed is 3600 rpm will be 0.5/3600 minutes = 8.3 milliseconds. 3. Transfer Rate. Transfer rate refers to the rate at which data are read from or written to the disk. Once the read/write head is positioned over the desired sector, the data are read/written at a speed determined by the rotational speed of the disk. If the rotational speed of a disk is 3600 rpm, the disk has 125 sectors/track, and 512 bytes/sector, then in one full revolution of the disk, the amount of data transferred will be 125 x 512=64,000 bytes = 64 k bytes (approximately). Hence the transfer rate of the disk system will be 64,000 * 3600/60 bytes/second = 38,40,000 bytes/second = 3.8 Megabytes/second (approximately). Notice that the transfer rate of a disk system depends on the density of the stored data and the rotational speed of the disk. Since the data transfer time is negligible (due to high transfer rate) as compared to seek time and latency, the average access time for a disk system is the sum of its average seek time and average latency. The average access time for different types of disk systems varies over a wide range and may be anywhere from 10 to 600 milliseconds. Since the access time for a piece of data stored on a disk depends on the physical location of the data, it is more correct to refer to a disk system as direct access storage device instead of random access storage device. Random access refers to a storage device in which the access time is independent of the physical location of the data. For example, primary storage i$ a random access storage. However, this distinction is not always observed strictly and hence disk systems are often referred to as random access storage devices. Disk Formatting Magnetic disks come in various different sizes. The size of a disk is usually referred to by its diameter. Typical disk sizes include those with 14 inch, 9 inch, 8 "inch, 5% inch, 3V4 inch, and VA inch diameter. Obviously, different size disks require different disk drives so that the dimensions of the mechanical components of the disk drive match with that of the disks to be used with it. Even for disks of the same size, all disk drives are not the same because we saw that different computer systems have different ways of organizing data on the disk surface. Hence disk drives of different computers may have their own way of defining tracks, sectors, and sector size (number of bytes/sector) to match their own way of organizing data. This implies that computer manufacturers should also manufacture the disks that can be used in their computer systems. This is a severe limitation because it prevents the use of disks manufactured by third party vendors into one's own computer system. To overcome this problem, the concept of disk formatting was introduced. According to this concept, before a disk can be used with a computer system, it must first be prepared by means of a process called disk formatting by the computer system. For this, the raw (unformatted) disk is inserted in the disk drive of the computer system and the disk formatting command is initiated. In this process, the disk
drive's read/write head lays down a magnetic pattern on the disk's surface. This pattern enables the disk drive to organize and store the data in the data organization defined for the disk drive. Most computers maintain on the disk a table with the sector and track locations of data. This table, known as the File Allocation Table (FAT), enables the computer to locate data easily. The disk formatting command also takes care of creating the FAT and setting aside sufficient space on the disk for it. Disk formatting is one of the basic tasks handled by the computer's operating system. If the formatting done by the disk drives of two different computer systems is exactly the same, the two computers are said to have compatible disk drives. Compatible disk drives allow disks prepared by one computer to be used by the other computer. This facilitates transfer of data and programs from one computer to another which are not linked together. Note that this is not possible by the use of disks between computers having incompatible disk drives. Disk Drive A magnetic disk has to be mounted on a disk drive before it can be used for reading or writing of information. A disk drive contains all the mechanical, electrical and electronic components for holding one or more disks and for reading or writing of information on to it. That is, it contains the central shaft on which the disks are mounted, the access arms assembly, the read/write heads, and the motors to rotate the disks and to move the access arms assembly. Although disk drives vary greatly in their shape, size and disk formatting pattern, they can be broadly classified into the following two types: 1. Disk drive with interchangeable magnetic disks. These disk drives allow the loading and unloading of magnetic disks as and when they are needed for reading/writing of data on to them. That is, at one instance a particular disk might be loaded for use, and at another instance another disk might be loaded in the same disk drive. This allows data on disks to be stored off-line and virtually unlimited capacity of the disk system because as many disks as required can be used one after another to store very large data sets. 2. Disk drive with fixed magnetic disks. These disk drives come along with a set of permanently fixed disks. In this case, the disks along with the read/write heads and the access mechanisms of the disk drive are permanently housed in sealed, contaminationfree containers. The sealed packaging allows the disks to operate in a dust-free environment. This enables the designers to provide increased data density of each square inch of disk surface in these disk drives by reducing the distance between the read/write head and the disk surface so that smaller magnetized spots can be precisely written and then retrieved. Thus these disk drives provide higher storage capacity with the same size disks and same number of disk surfaces. However, because of the sealed packaging, the disks are not removable from their disk drives. Hence the storage capacity of these disk systems is limited.
Disk Controller A disk drive is connected to and controlled by a disk controller which interprets the commands for operating the disk drive. Since disk is a direct access storage device, the disk controller typically supports only Read and Write commands. The disk address consisting of surface number, cylinder/track number, and sector number needs to be specified as parameters of the disk Read and Write commands. Often, a disk controller is connected to and controls more than one disk drive. In this case, the disk drive number must also be specified as a parameter of Read and Write commands so that the read/write operation is carried out on the specified disk surface of the correct disk drive. Types of Magnetic Disks All magnetic disks are round platters. They come in different sizes, different types of packaging, and can be made of rigid metal or flexible plastic. Based on these differences there are many different types of magnetic disks available today. However, all of them may be broadly classified into two types - floppy disks and hard disks. Floppy disks are individually packaged in protective envelopes or plastic cases, whereas hard disks may be packaged individually or in multiples in cartridges or contamination-free containers. Depending on the type of packaging, hard disks are further classified into Zip/Bernoulli disks, disk packs, and Winchester disks. The broad classification of magnetic disks is shown in Figure 8.18. The four types of magnetic disks are described below. Magnetic Disks Floppy Disks, Hard Disks Zip/Bernoulli, Disks, Disk Packs Floppy Disks A floppy disk is a round, flat piece of flexible plastic coated with magnetic oxide. It is encased in a square plastic or vinyl jacket cover. The jacket gives handling protection to the disk surface. Moreover, it has a special liner that provides a wiping action to remove dust particles that are harmful for the disk surface and the read/write head. Floppy disks are so called because they are made of flexible plastic plates which can bend, not hard plates. They are also known as floppies or diskettes. They were introduced by IBM in 1972 and are now being produced in various sizes and capacities by many manufacturers. Floppy-diskdrive A. floppy-disk drive is a device that is used to read/write data from/to floppy disks. The drive has a spindle that rotates the disk, and read/write heads that can move in and out to position the read/write heads on any track of the disk surface. It is of the interchangeable magnetic disks type. That is, it allows the loading and unloading of magnetic disks as and when they are needed for reading/writing data on to them. A floppy disk can be very easily loaded into or unloaded from a floppy-disk drive just by slipping the disk inside the drive's slot or by pressing the eject button on the drive. The disks are loaded into the disk drive along with the jacket cover and data are read/written through an aperture in the jacket (see Figure 8.19). Unlike the hard-disk drives, the read/write heads of a floppy-
disk drive make direct contact with the disk surface during the process of reading or writing. Hence floppy disks get worn out with constant use. Manufacturers guarantee about 2 million reads of a floppy disk track. In case of currently used floppy-disk drives, the rotational speed of a floppy disk is of the order of 300 to 400 rpm, and the data transfer rate is of the order of 10 to 30 Kilobytes/second. Types of floppy disks Although floppy disks and their corresponding drives are available in many sizes and capacities, the two types in widespread use today are the 5'/4-inch and the 3'/2-inch diskettes. They artrbriefly described below. 1. The 5%-inch diskette. It is a floppy disk of 5V4-inch diameter. The disk is encased in a square vinyl jacket that is harder than the disk itself but is still flexible. As shown in Figure 8.19, the jacket has on oval cutout (aperture) through which the read/write head makes direct contact with the disk surface. It moves radially along this opening of the jacket cover. The disk has a hole in the center in which the drive's spindle fits so that the disk can be rotated at high speed during operation. The disk rotates inside its jacket cover whose inner side is smooth and permits free rotation of the disk. Early 5'/4-inch floppy disks recorded data on only one side of the disk and were known as single-sided (SS) diskettes. However, today all 5'A-inch diskettes use both the disk surfaces for data recording and are known as double-sided (DS) diskettes. The doublesided 5%-inch diskette of today come in two capacities - double density (DD) and high density (HD). Recall that disk density refers to the number of bits that can be stored per square inch area on one surface of the disk. The DSDD (double-sided, double-density) 574-inch diskettes have 40 tracks, 9 sectors/track, and 512 bytes/sector, giving a total disk storage capacity of 2 (for two surfaces) x 40 x 9 x 512 = 3,68,640 bytes = 360 KB (approximately). On the other hand, the DSHD (double-sided, high density) 5V4-inch diskettes have 80 tracks, 15 sectors/track and 512 bytes/sector, giving a total disk storage capacity of 2 x 80 x 15 x .512=12,28,800 bytes = 1.2 MB (approximately). 2. The 3 ½ -inch diskette. It is a floppy disk of 3 '/2-inch diameter. The disk is encased in a square hard-plastic jacket cover. The jacket cover has a cutout (aperture) for the read/write head to make contact with the disk surface. However, unlike the 5'/4-inch diskettes, this aperture of the 3'/2-inch diskettes is covered with a sliding metal piece (see Figure 8.20). When the diskette is inserted into the drive, this covers slides back to expose the disk surface to the read/write head. Light indicator (blinks when read/write operation takes place) (a) Front view of a floppy disk drive. Opening for inserting the disk inside the drive Floppy disk eject button Write-protect plastic tab Sliding metal piece cover User's label for identification
(b) A 31/2-inch floppy disk All 3'/2-inch floppy disks are of double-sided type and record data on both the disk surfaces. However, they come in three different capacities - double density, high density, and very high density. The double-density 3'/2-inch diskettes have 40 tracks, 18 sectors/track, and 512 bytes/sector, giving a total disk storage capacity of 2 (for two surfaces) x40x 18x512 = 7,37,280 bytes = 720 KB (approximately). The high-density 31//2-inch diskettes have 80 tracks, 18 sectors/track, and 512 bytes/sector, giving a total disk storage capacity of 2 x 80 x 18 x 512 = 14,74,560 bytes =1.4 MB (approximately). The, very high-density 3'/2-inch diskettes have 80 tracks, 36 sectors/track, and 512 bytes/sector, giving a total disk storage capacity of 2 x 80 x 36 x 512 = 29,49,120 bytes = 2.88 MB (approximately). Figure 8.21 summarizes the capacities of the various types of 5'/4-inch and 3'/2-inch diskettes. Note that a smaller size disk can actually have more storage capacity than a larger size disk. Because of their hard-plastic jacket cover and the sliding metal cover over the aperture, the 3/4-inch floppy disks are more durable. The 3'/2-inch diskettes are slowly replacing the 5'A-inch diskettes because of their better durability, compact size (which can easily fit in a shirt's pocket), and higher capacity. As a result, the 5%-inch floppy disks are gradually disappearing. Size (Diameter in inches) 5'/4 5% VA VA VA Figure 8.21. diskettes.
No. of No. of No. of No. of Capacity in Approximate surfaces tracks sectors/track bytes/sector bytes capacity 2 2 2 2 2 Summary
40 9 512 80 15 512 40 18 512 80 18 512 80 36 512 of the capacities of the various types
3,68,640 12,28,800 7,37,280 14,74,560 29,49,120 of 5%-inch
360 KB 1.2 MB 720 KB 1.4 MB 2.88 MB and 31/4-inch
Note that the capacities given in Figure 8.21 are for DOS-based computers. As discussed before, different makes of computers may use different formatting style resulting in different capacities for the same disk. For example, a double-density 3'/2-inch diskette used on Macintosh systems has a storage capacity of 800 KB instead of 720 KB mentioned in Figure 8.21. Tips for proper usage and handling of floppy disks Although floppy disks are non-volatile storage media, data stored on them can still be lost or damaged if they are not handled properly. Here are a few tips for the proper usage and handling of floppy disks: 1. To prevent loss of stored data due to unintentional erasing/over-writing, the floppy disks should always be kept write protected when not in use. Fortunately, all floppy disks
have a simple write protection mechanism. For example, as shown in Figure 8.19, a 5%inch floppy disk can be write protected by covering the write-protect notch located near the upper-right corner of the disk covei with a write-protect sticker. These stickers are supplied together with the floppy disks when yov purchase them. Similarly, as shown in Figure 18.20, a 3%-inch floppy disk can be write protected b) sliding a plastic tab in the upper-left corner of the plastic cover such that a hole shows through the disk cover. 2. No sharp-edged device should be used for marking/writing something on the flexible type 5'/4-inch diskettes. A felt-tip pen should be used to write some information on the disk label and the writing should be done very gently without applying pressure. 3. A floppy disk should never be forced into or removed from a disk drive when the drive is in operation (its light indicator is on). Doing so may cause the read/write head to break, or may cause the read/write head to scratch the disk surface, resulting in damaging the disk or the data stored on it. 4. Their recording surfaces should never be touched by hand. These surfaces are already exposed through the oval cutout in case of 514-inch diskettes, and can be exposed by sliding the metal cover back by hand in case of 3'/2-inch diskettes. 5. The flexible type 514-inch diskettes should not be bent and should not be applied with any type of intense pressure. 6. To keep track of what information is stored on which floppy disk, and when was it last used, the floppy disks should always be properly labeled. 7. Since magnetic particles tend to lose their charge with time, the data stored on floppy disks may get damaged or lost after several months even if you take proper care of your floppy disks. This problem is known as byte rot. To prevent data loss due to byte rot problem, it is suggested that data stored on a disk should be refreshed (re-written) every year. To do this, copy the contents of an entire floppy disk on to the hard disk of a computer, and then copy them back on to the same floppy disk. 8. They should always be stored away from anything that has a magnetic charge or that generates a magnetic field. For example, they should be kept away from magnets, magnetic materials, and electric equipment. 9. They should always be stored in a place where it is not very hot or humid. Floppy disks being made of thin plastic may get bad due to warping under extreme temperature and humidity conditions. 10. Multiple disks should not be stacked on top of each other. They should be stored by keeping them standing against their edge in specially designed plastic boxes to prevent them from dust particles. Floppy disks are very cheap as compared to other secondary storage devices. They are also a convenient off-line storage medium for the small computer users. Hence diskettes
are currently one of the most popular, inexpensive secondary storage medium used in small computer systems. Hard Disks Hard disks are the primary on-line secondary storage device for most computer systems today. Unlike floppy disks, which are made of flexible plastic or mylar, hard disks are made of rigid metal (frequently aluminium). The hard disk platters come in many sizes ranging from 1 to 14-inch diameter. Types of hard disks Depending on how they are packaged, hard disks are normally categorized into the following three types: 1. Zip/Bernoulli Disks. In this type, a single hard disk platter is encased in a plastic cartridge. A commonly used zip disk is of 3'/2 inch size having a storage capacity of about 100 MB depending on the formatting style used by a particular computer system. Its disk drive is called a zip drive. A zip drive may be of portable or fixed type. The fixed type is a part of the computer system, permanently connected to it. The portable type can be carried to a computer system, connected to it for the duration of use, and then can be disconnected and taken away when the work is done. The zip disks can be easily insertedanto or removed from a zip drive just as we insert and remove floppy disks in a floppy disk drive or a video cassette in a VCR. 2. Disk Packs. A disk pack consists of multiple (two of more) hard disk platters mounted on a single central shaft. Thus all the disks of a disk pack revolve together at the same speed. As mentioned before, the disk drive of a disk pack has a separate read/write head for each disk surface excluding the upper surface of the topmost disk and the lower surface of the bottommost disk. These two surfaces are not used for data recording in a disk pack. When not in use, disk packs are stored in plastic cases as shown in Figure 8.22. They are of removable/interchangeable type in the sense that they have to be mounted on the disk drive before they can be used, and can be removed and kept offline when not in use. That is, different disk packs can be mounted on the same disk-pack drive at different instances of time. This gives virtually unlimited storage capacity to disk packs. 3. Winchester Disks. A Winchester disk also consists of multiple (two or more) hard disk platters mounted on a single central shaft. However, the main difference between a Winchester disk and a disk pack is that Winchester disks are of fixed type. That is, the hard disk platters and the disk drive are sealed together in a contamination-free container and cannot be separated from each other. Thus as opposed to disk packs which have virtually unlimited capacity, Winchester disks have limited capacity. However, for the same number of disk platters of the same size, Winchester disks can manage, to have larger storage capacity than disk packs due to the following reasons: - Because both the disk platters and the disk drive are sealed in a contamination-free container and do not require to be separated later, all the surfaces of all the disk platters
(including the upper surface of the topmost platter and the lower surface of the bottommost platter) are used for data recording in case of Winchester disks. That is, for a Winchester disk with foui platters, there are eight surfaces on which data can be recorded, as opposed to six surfaces in case of a disk pack with four platters. - The contamination-free environment allows Winchester disks to employ much greater precision of data recording and accessing, resulting in greater density of data storage than the interchangeable disk packs. Winchester disks were so named after the 30-30 Winchester rifle because the early Winchester disk systems had two 30-MB disks sealed together with the disk drive. The storage capacity of today's Winchester disks is usually of the" order of a few tens of megabytes to a few gigabytes (10 bytes). Why do hard disks outperform floppy disks? Hard disks outperform floppy disks in both data storage capacity and data access speed due to the following reasons: 1. A floppy disk drive is normally designed to operate with only one floppy disk platter, whereas a hard disk drive normally operates with multiple hard disk platters, providing several times more data storage capacity. 2. The rigid construction of hard disks as opposed to flexible construction of floppy disks, allows higher precision in pinpointing positions on the disk, resulting in higher data storage densities on disk surfaces. The contamination-free sealed environment of Winchester disks allows its read/write head to fly much closer to the surface of the disk, allowing much greater precision and further increase in disk densities. For instance, floppy disks normally have 135 tracks per inch, whereas hard disks can easily have more than 1000 tracks per inch. 3. The rigid construction of hard disks also allows them to spin at much greater speeds than floppy disks, resulting in faster data access. Floppy disks tend to bend at higher speeds due to their flexible construction. For example, a floppy disk normally spins at 360 rpm, whereas hard disks spin at 3600 rpm, 5400 rpm, or even 7200 rpm. 4. In most hard disk drives, the disk(s) keep spinning continuously. In contrast, the diskette of a floppy disk drive is set in motion only when a command is issued to read from or write to the disk. This also contributes to faster data access for hard disks as compared to floppy disks. Advantages and Limitations of Magnetic Disks Advantages As a secondary storage device, magnetic disks have the following advantages: 1. Unlike magnetic tapes, which support sequential access of data, magnetic disks support direct access of data. Hence they are more suitable for a wider range of applications. 2. Due to its random access property, magnetic disks are often used simultaneously by multiple users as a shared device. For example, Winchester disks and disk packs are
often used as on-line secondary storage devices, in which case they store data of multiple users of the computer system. A tape is not suitable for such type of usage due to its sequential-access property. 3. Magnetic disks are suitable for both on-line storage of date. For example, Winchester disks and disk packs are often used as on-line secondary storage devices, whereas floppy disks and zip disks are used as off-line secondary storage devices. In fact, the high-capacity Winchester disks have made it possible for today's most personal computer users to enjoy the convenience of having all data and software readily accessible at all times. 4. Except for the fixed type Winchester disks, the storage capacity of other magnetic disks is virtually unlimited because as many disks are required can be used for storing very large data sets. 5. Due to their low cost and high data recording densities, the cost per bit of storage is low for magnetic disks. An additional cost benefit is that magnetic disks can be erased and reused many times. 6. Floppy disks and zip disks are compact and light in weight. Hence they are easy to handle and store. Very large amount of data can be stored in a small storage space. 7. Due to their compact size and light weight, floppy disks and zip disks are also easily portable from one place to another. They are often used for transferring data and programs from one computer to another, which are not linked together. 8. Any information desired from a disk storage can be accessed in a few milliseconds because it is a direct access storage device. This is not possible in case of a tape storage, which is a sequential access storage device. 9. Data transfer rate for a magnetic disk system is normally higher than a tape system. 10. Magnetic disks are less vulnerable to data corruption due to careless handling or unfavorable temperature and humidity conditions than magnetic tapes. Limitations As a secondary storage device, magnetic disks have the following limitations: 1. Although magnetic disks may be used for both applications requiring random processing of data and applications requiring sequential processing of data, for applications of the latter type, use of magnetic disks may be less efficient than magnetic tapes. 2. It is more difficult to maintain the security of information stored on magnetic disks which are used as shared, on-line secondary storage devices, as compared to information stored on magnetic tapes or on other types of magnetic disks. 3. For Winchester disks, a disk crash or drive failure often results in the loss of entire data stored on it. It is not easy to recover the lost data. That is why suitable backup procedures are suggested for data stored on Winchester disks. 4. Some types of magnetic disks, such as disk packs and Winchester disks, are not so easily portable like magnetic tapes. 5. On a cost-per-bit basis, the cost of magnetic disks is low, but the cost of magnetic tapes is even lower. 6. They must be stored in a dust-free environment.
7. Floppy disks, zip disks and disk packs should be labeled properly to prevent erasure of useful data by mistake. Uses of Magnetic Disks Magnetic disks are typically used for one or more of the following purposes: 1. For applications that are based on random data processing. 2. As a shared on-line secondary storage device. Winchester disks and disk packs are often used for this purpose. 3. As a backup device for off-line storage of data so that if by accident, the data on an on-line storage device gets corrupted, it can be retrieved from the backup storage. Floppy disks, zip disks, and disk packs are often used for this purpose. 4. Archiving of data which are not used frequently, but which may be used once in a while. Floppy disks, zip disks, and disk packs are often used for this purpose. 5. Transferring of data and programs from one computer to another, which are not linked together. Floppy disks and zip disks are often used for this purpose. 6. Distribution of software by vendors. Originally sold software or software updates are often distributed by vendors on floppy disks and zip disks. OPTICAL DISK As compared to magnetic tape and magnetic disk, optical disk is a relatively new secondary storage medium. During the last few years, it has proved to be a promising random access medium for high capacity secondary storage because it can store extremely large amounts of data in a limited space. An optical-disk storage system consists of a rotating disk which is coated with a thin metal or some other material that is highly reflective. Laser beam technology is used for recording/reading of data on the disk. Due to the use of laser beam technology, optical disks are also known as laser disks or optical laser disks. Basic Principles of Operation Storage Organization Unlike magnetic disks that have several concentric tracks, an optical disk has one long track that starts at the outer edge and spirals inward to the center (see Figure 8.23). This spiral track is ideal for reading large blocks of sequential data, such as music. But it makes for slower random access time than the concentric tracks used by magnetic disks, whose sectors can be located faster because they are always found on a given track at a fixed distance from the center. Like a track on a magnetic disk, the track of an optical disk is split up into sectors, but with optical disks each sector has the same length regardless of whether it is located near the disk's center or away from the center. This type of data organization allows data to be packed at maximum density over the entire disk. However, it also requires a more
complicated drive mechanism because the rotation speed of the disk must varv inversely with the radius; the drive must slow down the disk's rotation speed to read sectors towards the outside of the disk and speed it up to read sectors towards the center of the disk. (a) Track pattern on an optical disk (b) Track pattern on a magnetic disk Storage Capacity The cost-per-bit of storage is very low for optical disks because of their low cost and enormous storage density. They come in various sizes ranging from 12-inch to 4.7-inch diameter. The most popular one is of 5.25 inch diameter whose capacity is around 650 Megabytes. This storage capacity is equivalent to about 2, 50,000 pages of printed text or total capacity of 550 double-sided, high-density floppy disks of the same size. This tremendous storage capacity makes optical disks the most suitable storage media for multimedia applications (described in Chapter 19). As optical disks have a single track, their storage capacity depends on the following: 1. Number of sectors, and 2. Number of bytes per sector It is a multiple of the above two parameters. That is, Storage capacity of an optical disk = Number of sectors x Number of bytes per sector The 5.25-inch optical disks typically have 3,30,000 sectors, each of 2352 bytes. That gives a total capacity of 3,30,000 x 2352 = 776 x 106 bytes = 776 Megabytes. This is the unformatted capacity. The formatted capacity is about 650 Megabytes. Access Mechanism Optical disks use laser beam technology for recording/reading of data on the disk surface. With this technology, the read/write head used in magnetic storage is replaced by two laser beam sources. One laser beam (of greater intensity) is used to write to the recording surface by etching microscopic pits on the disk surface, and another laser beam (of lesser intensity) is used to read the data from the light-sensitive recording surface. That is, data recording is done by focusing the stronger laser beam on the surface of the spinning disk. The laser beam is turned on and off at a varying rate because of which tiny pits (visible only through a powerful microscope) are burnt into the metal coating of the disk along its tracks. To read the stored data, the less-powerful laser beam is focused on the disk surface. As shown in Figure 8.24, this beam is strongly reflected by the coated surface (known as a land) and weakly reflected by the burnt surface (known as a pit), producing patterns of on and off reflections that are converted into electronic signals of binary Is and 0s by a sensor. Note that a mechanical read/write access arm is not needed in this case because a light beam can be easily deflected to the desired place on the optical disk.
Access Time Since a mechanical read/write access arm is not needed, the delay that occurs in case of magnetic disks due to mechanical movements of the access arm will not be there in case of optical disks. Hence you may think that data accessing from optical disks will be much faster (access time will be less) as compared to that from magnetic disks. However this is not true. Optical disk drives are slower (have larger access time) than magnetic disk drives due to the following reasons: 1. We saw that the sectors of an optical disk are arranged on a continuous spiral track. This data organization results in a slower random access time than the concentric tracks used by magnetic disks, whose sectors can be located faster because they are always found on a given track at a fixed distance from the center. 2. We also saw that with optical disks each sector has the same length regardless of whether it is located near the disk's center or away from the center. This data organization requires a more complicated drive mechanism because the rotation speed of the disk must vary inversely with the radius; the drive must slow down the disk's rotation speed to read sectors towards the outside of the disk and speed it up to read sectors towards the center of the disk. This access mechanism leads to slower data access (larger access time) as compared to magnetic disks, in which case the disks rotate at a constant speed irrespective of the location of the data to be accessed. Access times for optical disks are typically in the range of 100 to 300 milliseconds. Compare this with the access times of floppy disks which are typically in the range of 100 to 200186 Optical Disk Drive An optical disk has to be mounted on an optical disk drive before it can be used for reading or writing of information. An optical disk drive contains all the mechanical, electrical and electronic components for holding an optical disk and for reading or writing of information on to it. That is, it contains the tray on which the disk is kept, the read/write laser beams assembly, and the motor to rotate the disk. A typical optical disk drive is shown in Figure 8.25. An optical disk drive differs from a magnetic disk drive in the following manner: 1. As shown in Figure 8.24, it uses laser beam technology for data reading/writing and has no mechanical read/write access arm. 2. Magnetic disk drives use a constant angular velocity (CAV) encoding scheme whereas optical disk drives use a constant linear velocity (CLV) encoding scheme. That is, in case of hard disk drives the disk spins at a constant rate of speed and each sector occupies the area subtended by a fixed angle. On the other hand, in case of optical disk drives, the rotational speed of the disk varies inversely with the radius; the motor slows down the disk to read sectors towards the outside of the disk and speeds it up to read sectors towards the inside of the disk. Doing this requires a more complicated drive mechanism that can overcome the inertia of the disk when accelerating and the momentum when decelerating, which slows the drive's random access time. But this encoding scheme also allows data to be packed at maximum density over the entire disk.
Types of Optical Disks All optical disks are round platters. They come in different sizes and capacities. The two most popular types of optical disks in use today are CD-ROM and WORM disks. They are described below. CD-ROM CD-ROM stands for Compact Disk-Read-Only Memory. It is a spin off of music CD technology and works much like the music CDs used in music systems. In fact, if you have a sound board and speakers connected to your computer, you can play music CDs with your PC. The CD-ROM disk is a shiny, silver color metal disk of 5% inch (12 cm) diameter. It has a storage capacity of about 650 Megabytes. It is so called because of its enormous storage capacity on a compact-size disk, and because it is a read-only storage medium. That is, these disks come pre-recorded and the information stored on them cannot be altered. CD-ROM disks are pre-stamped by their suppliers. The process of recording of information on these disks is called mastering. That is, CD-ROM disks are created at a mastering facility, just as music CDs are created at a recording studio. For this, a master disk is first- made with the high-power laser, which is then replicated on multiple disks by a stamping process (disk pressing process). In this process, the embossed surface of pits and lands is accurately reproduced on each copy. The replicated copy is then plated with a reflective surface, and a clear protective layer is added to complete the replication process. These copies are then distributed with their pre-recorded contents. Once inserted into the disk drive, the information stored on the disk can be read into the primary storage of the computer system for processing or display. However, the data on the disk are fixed and cannot be altered. This is in contrast to the read/write capability of magnetic tapes and magnetic disks. The fact that a normal user cannot write data to a CD-ROM disk does not mean that this storage medium is not useful. In fact, there are many applications which make use of huge volumes of data that rarely change, For example, dictionaries; encyclopedias; medical, legal, or other professional reference libraries; music; and video can be stored electronic form on a CD-ROM disk for efficient, electronic access by a computer user because all these require tremendous amount of data that a normal user needs to only refer to (read). Because of their high storage capacities and the fact that no one can change the data stored on them, CD-ROM disks provide an excellent medium to distribute large amounts of data at low cost. For example, infopedia is a single CD-ROM disk that holds a complete encyclopedia, or a dictionary, or a world atlas, or a dictionary of quotations or biographies of great people, or information about all educational institutions of a particular type in a country etc CD-ROM disks are nowadays also used extensively for distribution of electronic version of conference proceedings, journals, magazines, books, etc. They are also the distribution medium of choice of many software companies who distribute every new version and upgrade of their software products to their clients on CD-ROM disks. They are also extensively used for storage and
distribution of multimedia applications which involve the integration of text, sound, graphics, motion video, and animation. To allow CD-ROM disks to be widely distributed and used on different computers, a standard for recording information on CD-ROM disks, called ISO 9660, has been evolved by the International Standards Organization According to this standard, a Volume Table of Contents (VTOC) is created on the disk which allows storage and direct access of 1,40,000 files on the disk. With this, any one of the 1,40,000 files can be opened with a single seek. A software known as MSCDEX (Microsoft CD Extension), which has been developed by Microsoft Corporation, is used for recording data in ISO 9660 format on CD-ROM disks. The ISO 9660 format is accented by all CD-ROM vendors. WORM Disk We saw that CD-ROM disks are produced by manufacturers with expensive duplication equipment This is economical only when a large number of copies of a master disk are to be produced. Often users of computer systems would like to produce a few copies, or a single copy of their own information in-house. This is possible by the use of WORM disks. WORM stands for write-once, read-many. WORM disks allow the users to create their own CD-ROM disks by using a CD-recordable (CD-R) drive, which can be attached to a computer as a regular peripheral device. WORM disks, which look like standard CDROM disks, are purchased blank and encoded using a CD-R drive. The information recorded on a WORM disk by a CD-R drive can be read by any ordinary CD-ROM drive. As the name implies, data can be written only once on a WORM disk, but can be read many times. That is, as with a CD-ROM disk, once data have been etched on to the surface of a WORM disk, they become permanent that can be read but never altered. Moreover, writing on a WORM disk cannot be done in multiple sessions, and all the data to be recorded have to be written on the disk surface in a single recording session. The same laser beam technology as discussed before is used for recording and reading of data. Because of their permanent recording nature and large storage capacity, WORM disks are the preferred storage media for archival application, which now depend on microfilm, magnetic, or paper file media. Data archiving enables old data to be recorded on a secondary storage media (such as a WORM disk) so that they can be deleted from the online storage device (such as a hard disk), thus freeing space for new data. WORM disks are also an ideal medium for making a permanent record of data. For example, many banks use them for storing their daily transactions. The transactions are written to a WORM disk in-house by the bank and become a permanent record that can be only read but never altered. Advantages and Limitations of Optical Disks Advantages
As a secondary storage device, optical disks have the following advantages: 1. The cost-per-bit of storage for optical disks is very low because of their low cost and enormous storage density. 2. The use of a single spiral track makes optical disks an ideal storage medium for reading large blocks of sequential data, such as music. 3. Optical disk drives do not have any mechanical read/write heads to rub against or crash into the disk surface. This makes optical disks a more reliable storage medium than magnetic tapes or magnetic disks. 4. Optical disks have a data storage life in excess of 30 years. This makes them a better storage medium for data archiving as compared to magnetic tapes or magnetic disks. 5. Since data once stored on an optical disk becomes permanent, the danger of stored data getting inadvertently erased/overwritten is not there with optical disks. 6. Due to their compact size and light weight, optical disks are easy to handle, store, and port from one place to another. 7. Music CDs can be played on a computer having a CD-ROM drive along with a sound board and speakers. This allows computer systems to be also used as music systems whenever desired. Limitations As a secondary storage device, optical disks have the following limitations: 1. It is a read-only (permanent) storage medium. Data once recorded, cannot be erased and hence the optical disks cannot be reused. 2. The data access speed for optical disks is slower than magnetic disks. 3. Optical disks require a more complicated drive mechanism than magnetic disks. Uses of Optical Disks Optical disks are typically used for one or more of the following purposes: 1. For distributing large amounts of data at low cost. For example, a complete encyclopedia, dictionary, world atlas, dictionary of quotations, biographies of great people, information about all educational institutions of a particular type in a country, etc. are often distributed on CD-ROM disks. 2. For distribution of electronic version of conference proceedings, journals, magazines, book, product catalogs, etc. 3. For distribution of new or upgraded versions of software products by software vendors. 4. For storage and distribution of a wide variety of multimedia applications. 5. For archiving of data, which are not used frequently, but which may be used once in a while. 6. WORM disks are often used by end-user companies to make permanent storage of their own proprietary information. For example, many banks use them for making a permanent record of their daily transactions.
MAGNETO-OPTICAL DISK We saw that both CD-ROM and WORM optical disks are permanent storage devices. That is, data once written on them cannot be erased and overwritten. Hence unlike magnetic tapes and magnetic disks, optical disks are not reusable. To overcome this limitation of optical disks, researchers invented magneto-optical disk technology which integrates optical and magnetic disk technologies to enable rewritable storage with laserbeam technology. Since magneto-optical disks are rewritable, they are also known as rewritable optical disks. A magneto-optical disk consists of a rotating disk that is covered with magnetically sensitive metallic crystals sandwiched inside a thin layer of plastic coating. In its normal state, the plastic coating is solid, preventing the metallic crystals from moving. To write data on the disk, the laser beam of greater intensity is focused on the surface of the disk at the location of writing. The intensity of this laser beam is just strong enough to melt the plastic coating for a very short instance of time that is just long enough to allow a magnet to change the orientation of the metaliic crystals. The magnet has an effect at only the point where the laser beam is focused because the plastic coating at only that point is fluid enough to allow the metallic crystals to be reoriented. To read data from a magneto-optical disk, the laser beam of lesser intensity is focused on the disk track formed by the metallic crystals. As the disk spins and the metallic crystals pass under the laser beam, depending on their orientation (alignment), some metallic crystals reflect the light while others scatter the light, producing patterns of on and off reflections that are converted into electronic signals of binary Is and Os by a sensor. Magneto-optical disks are today available in various sizes and storage capacities. The ones having 5'/4-inch and 3/2-inch diameter have storage capacities in excess of 1 GB. However, the access speed of magneto-optical disk drives is slower than hard-disk drives. The access time for magneto-optical disk drives is about 30 milliseconds which is almost as fast as the access speed of slower hard-disk drives. Moreover, magneto-optical disk drives are more expensive and less feliable than magnetic disk drives. Due to these reasons, magneto-optical disks are not yet as popular as magnetic disks or optical disks. But as magneto-optical disk technology matures to offer reliable and cost-effective rewritable storage media, it eventually may dominate the secondary storage market in the future. Magneto-optical disks are ideal for multimedia applications that require large storage capacity. They are also used for applications that require large volumes of storage with relatively little update activity, and for applications that require a reusable, portable and durable secondary storage device. MASS STORAGE DEVICES Mass storage devices are storage systems that have storage capacity of several trillions of bytes of data.
Researchers are constantly trying to improve the secondary storage media discussed above by increasing their capacity, performance, and reliability. However, any physical storage media has a limit to its capacity and performance. When it gets harder to make further improvements in the physical storage media and its drive, researchers normally look for other methods for improvement. Mass storage devices are a result of such improvements which researchers have resorted to for larger capacity secondary storage devices. The idea is to use multiple units of the storage media being used (such as multiple disks, multiple tapes, multiple CD-ROMs, etc.) as a single secondary storage device. Based on this idea, the three types of commonly used mass storage devices are: 1. Disk array, which uses a set of magnetic disks. 2. Automated tape library, which uses a set of magnetic tapes. 3. CD-ROM jukebox, which uses a set of CD-ROMs. These three types of mass storage devices are briefly described below. The average access times of mass storage devices are measured in seconds instead of milliseconds because a transport mechanism must first move to retrieve the storage media (such as disk, tape or CD-ROM) upon which the desired data is stored. It requires several seconds to locate the storage media specified and then few milliseconds are needed to transfer the data to the memory. Thus an average access time of few seconds is common for these storage devices. But a mass storage device has huge storage capacity and a very small cost per bit stored. Relatively slow access times limit the use of mass storage devices in many applications. However, mass storage devices are cost-effective alternative to on-line magnetic tape or disk storage in applications that require huge storage capacity and in which rapid access to data is not essential. When used for off-line storage, mass storage devices are often referred to as archival storage because of the very large volumes of historical or backup data that they can store. Disk Array A disk array, which first appeared on the market in 1993, is a set of hard disks and hard disk drives with a controller mounted in a single box. All the disks of a disk array form a single large storage unit. A disk array is commonly known as a RAID (Redundant Array of Inexpensive Disks). The term inexpensive comes from the fact that each of the medium-sized hard disks in the disk array is much less expensive than a single large hard disk. Although RAID systems were originally developed to provide large secondary storage capacity with enhanced performance, today they are also becoming popular due to enhanced reliability. How do RAID systems provide enhanced storage capacity, enhanced performance, and enhanced reliability is briefly explained below. Suppose we have a 4 Gigabyte disk whose each track contains 128 sectors, each of 512 bytes. The system uses 4 sectors per cluster (also known as a block), so that every read/write access to the disk transfers 2 Kilobytes (512 bytes x 4) of data. The disk makes one revolution in 10 ms and can transfer a block in 0.15 ms. Now suppose a RAID system uses 4 of these disks and puts them in the same enclosure. This 4-disk RAID unit will have total storage capacity of 4 x 4 = 16 Gigabytes. This naturally results in a four
fold increase in storage capacity as compared to a single disk's storage capacity. Furthermore, this RAID unit will have blocks of 4 x 2 = 8 Kilobytes which will consist of the matching blocks (the ones with the same block number) on the four disks (recall the cylinder concept of data organization when multiple disks are used in a disk pack). The access time for a block on this RAID unit will be the same as for the individual disks, but the transfer rate will be four times since we can now deliver an 8 Kilobyte block in 0.15 ms. The transfer rate has increased from 2K/0.00015 = 13.33 Megabytes/second to 8K/0.00015 = 53.33 Megabytes/second. Clearly the storage capacity and performance can be further enhanced by using even more disks in the RAID with parallel data transfer. Enhanced reliability is achieved by RAID units by taking advantage of the presence of multiple disks. The two commonly used techniques used for this purpose are mirroring and striping. In case of mirroring, the system makes exact copies of files on two hard disks. That is, one is the mirror image of the other one and if one of the two copies gets corrupted, the other copy is available for accessing. In case of striping, a file is partitioned into smaller parts and different parts of the file are stored on different disks. In this case, a spare disk is used to record parity data that are used for error checking and data recovery in case of a disk failure. When the file is read, the data on the spare disk ensures that the data read from the other disks are correct. In addition to enhanced reliability, both mirroring and striping also speed up the process of reading data from the disk array because the different hard disk drives of the RAID unit can read different parts of the file at the same time. Figure 8.26 shows a RAID unit having 10 disks. A large RAID unit can be composed of more than one hundred disks and can have storage capacity of several terabytes (1 terabyte = 1012 bytes). Note that the same RAID unit can be used in several different ways (combinations of striping, mirroring, and error detection and correction) by changing the control software. Depending on the various types of control software, there are the following commonly accepted levels of RAID systems, each one with unique economic, performance and reliability tradeoffs: 1. 2. 3. 4. 5. 6. 7.
RAID level 0: It uses striping. RAID level 1: It uses mirroring. RAID level 2: It uses striping plus error detection and correction. RAID level 3: It uses striping with dedicated parity. RAID level 4: It uses modified striping with dedicated parity. RAID level 5: It uses modified striping with interleaved parity. RAID level 0 + I: It uses striping and mirroring.
With such a wide range of reliability features inbuilt in them, modern RAID systems have mean time between data loss (MTBL) of well over a million hours, tlence they are often considered as a basic unit for building reliable computer systems. With more and more organizations relying on information technology to run their business, unscheduled
interruptions have more serious consequences. As a result, today many organizations are turning to the RAID as a safer way to store large amounts of data. Automated Tape Library An automated tape library is a set of magnetic tapes with a controller mounted in a single unit. The unit has one or more tape drives to read and write data on the tapes in the tape library. In case of units having multiple tape drives, it is possible to read/write data from/to multiple magnetic tapes simultaneously resulting in faster data transfer rates. Multiple drives also lead to improved reliability of the storage unit because if one of the drives fails, the unit can continue to function with other drives at a slower speed. The unit also has robotic arms to retrieve the appropriate tape from the tape library and mount it on one of the tape drives for processing. The tape is automatically returned to the library at the end of the job. A large tape library can accommodate up to several hundred high capacity magnetic tapes bringing the storage capacity of the storage unit to several terabytes (1 terabyte is equal to one trillion bytes). Automated tape libraries are typically used for data archiving and as an on-line data backup device for automated backup. They are mainly found in large computer centres. CD-ROM Jukebox The principle of operation of a CD-ROM jukebox is very similar to that of an automated tape library except for the fact that it uses a set of CD-ROMs instead of magnetic tapes as the storage media. That is, a CD-ROM jukebox is a set of CD-ROM disks with a controller mounted in a single unit. The unit has one or more CD-ROM drives to read data from the CD-ROM disks in the jukebox. In case of units having multiple CD-ROM drives, it is possible to read data from multiple CD-ROM disks simultaneously resulting in faster data transfer rates. Multiple drives also lead to improved reliability of the storage unit because if one of the drives fails, the unit can continue to function with other drives at a slower speed. The unit also has robotic arms to retrieve the appropriate CD-ROM and mount it on one of the CD-ROM drives for processing. The CD-ROM is automatically returned to the appropriate slot at the end of the job. A large CD-ROM jukebox can accommodate up to several hundred CD-ROM disks bringing the storage capacity of the storage unit to several terabytes. CD-ROM jukeboxes are typically used for archiving read-only data that needs to be accessed on-line but where access time is not very critical. For example, on-line museums, on-line digital libraries and on-line encyclopedia are a few systems that may make use of this mass storage device in an effective manner. OTHER RELATED CONCEPTS Few other concepts related to secondary storage of computer systems are briefly described below.
Drive Interface Standards Secondary storage devices need a controller to act as an intermediary between the device and the rest of the computer system. On some computers, the controller is an integral part of the computer's main motherboard. On others, the controller is an expansion board that connects to the system bus by plugging into one of the computer's expansion slots. In order that devices manufactured by independent vendors can be used with different computer manufacturers, it is important that the controllers follow some drive interfacing standard. The four commonly used drive interface standards are: 1. Integrated Drive Electronics (IDE). IDE incorporates much of the intelligence of the controller into the drive itself. Because most of their controlling electronics are on the drive itself, IDE controllers are small, simple cards that do little more than providing an IDE drive with a place to plug into the computer's bus. In fact, some computer manufacturers build IDE support right into their motherboards. IDE provides datatransfer rates of about 1 Megabytes per second under ideal conditions. 2. Enhanced Small Device Interface (ESDI). Like IDE, ESDI (pronounced "es-dee") also incorporates much of the intelligence of the controller into the drive. This provides a simpler interface with the computer and leads to faster and more reliable operation. An ESDI controller can theoretically address up to 1 terabyte (1012 bytes) of disk space and can reliably transfer up to 3 Megabytes per second. 3. Intelligent Peripheral Interface-level 3 (IPI-3). IPI-3 is used in minicomputers and mainframe systems for interfacing faster, larger disks (having diameter of 8 inches or more) with the main computer system. It allows faster data transfer from the disk to the computer's memory as compared to IDE or ESDI. 4. Small Computer System Interface (SCSI). In case of IDE, ESDI and IPI-3, a separate device controller is required in the computer system for each type of device that has to be connected to it. SCSI (pronounced "scuzzy") is an interesting and important variation of the separate device controller idea for each device. It uses a generic device controller (called SCSI controller) on the computer system and allows any device with an SCSI interface to be directly connected to the SCSI bus of the SCSI controller. The SCSI interface of a device contains all circuitry that the device needs to operate with the computer system. As shown in Figure 8.27, a SCSI controller connects directly to the computer bus on one side and controls another bus (called SCSI bus) on the other side. Since the SCSI controller is connected to the computer's bus on one side and to the SCSI bus on the other side, it can communicate with the processor and memory and can also control the devices connected to the SCSI bus. The SCSI bus is a bus designed for connecting devices to a computer in a uniform way. In effect, the SCSI controller extends the computer's bus outside the main system box and allows other devices to be connected to the computer with the help of this extended bus. The important thing about the SCSI bus is that there is
a standard protocol for sending device control messages and data on the bus. Each device connected to the SCSI bus has a standard SCSI interface (a single chip) that knows about the bus protocol. Since SCSI extends the computer's bus outside the main computer box, it allows any SCSI device to be connected to the computer without the need to open the computer box and install a device controller on its internal bus. Bringing the computer's bus directly into the drive allows even higher data-transfer rates than are possible with ESDI. A SCSI being an extension of a computer's bus, it can accommodate multiple devices simultaneously. Most current implementations of SCSI limit the number of devices to seven due to the bus capacity limitation. Moreover, SCSI being an extension of the bus, any type of device (hard-disk drive, optical-disk drive, tape drive, printer, plotter, scanner, etc.) can be linked (or daisy-chained) on a single SCSI port (see Figure 8.27). As the name implies, SCSI started out as a standard interface for connecting devices to small computer systems. But the advantages of a common interface soon became apparent in all sizes of systems, and newer versions of SCSI have much higher datatransfer rates and are used in high-end systems. Data Backup What is Data Backup? Data backup is the process of creating a copy of the data from an on-line storage device to a secondary storage backup device such as, magnetic tape, floppy disk, zip disk, disk pack or WORM disk, (not necessarily but usually for off-line storage of the data) so that if by accident the data on the on-line storage device gets corrupted or lost, it can be retrieved from the backup device and stored back on the on-line storage device. Why Backup Data? Taking periodic backup of on-line data and keeping them on off-line secondary storage devices is an important and necessary practice for computer users because the data on an on-line storage device, such as a hard disk, can be damaged or lost in any one or more of the following ways: 1. A disk crash. 2. A virus attack. 3. A hardware malfunction. 4. An unintended accidental deletion of useful files. 5. A natural disaster like fire or earthquake damaging the computer system. The amount of time it would take you to re-create all of your files if you lost them is far greater than the few minutes it will take you to back them up. Hence protecting valuable data with regular backups is a must for all computer centres and all computer users. Types of Backup
The two basic types of backup are full backup and incremental backup. They are described below. 1. Full Backup. In this case, all the data on the on-line secondary storage device is copied on the backup device at the time of backup. That is, the backup procedure creates a copy of every file from the on-line secondary storage device on to the backup device, whether the file has been changed or not. Depending on the volume of data being backed up and the backup device used, full backups can take anywhere from a few minutes to a few hours to complete. 2. Incremental Backup. Since backup is an expensive process both in terms of the time taken to complete and the storage space occupied on the backup device, full backup is not always suggested. To save on backup time and storage space, one can often go for incremental backup instead of full backup. In this case, only the newly created files or files that have been changed since the last backup are copied on the backup device from the on-line secondary storage device at the time of backup. The obvious question is how to keep track of the newly created and modified files since the last backup. Fortunately, every file stored in the system contains the date when it was created or last changed, so the backup software (known as backup utility) can simply compare these dates to those in its log for taking incremental backup. Incremental backup is faster than full backup because it needs to create a copy of only the new and modified files since the last backup. This may actually result in copying of only a few files out of hundreds of files stored on the on-line secondary storage device. However, full backup is simpler and a little safer when the data has to be restored back from the backup device on to the on-line secondary storage device. If you want to restore a particular file that was lost, you need to search for it at only one place in case of full backup, whereas in case of incremental backup you need to search for it on several backups tracing back from the most recent backup till you find the file. Similarly, in the event of a disk crash, when all the files are lost and need to be restored back on to a new disk, it is very easy to do this with full backup as compared to incremental backup because the full backup is a mirror image of the entire disk and can be restored as it is. Backup Policy Every computer system needs to have a well defined backup policy. The backup policy for a computer system depends on its nature of usage, types of users, its operating environment, etc. A backup policy normally decides about the following: 1. What is the periodicity of backup? That is, is the backup procedure executed hourly, daily, weekly, fortnightly, monthly, etc. The periodicity normally depends on the nature of usage and the criticality of the stored data of a system. 2. Whether to take full or incremental backup? This again depends on the nature of usage of the system. For example, if a user creates only a few files (say letters) a day, an incremental backup performed at the end of each day or even once a week may be sufficient. On the other hand, if a user frequently creates many files, weekly full backups
are safer. A compromise that works well for many systems is performing incremental backups daily and full backups once in a week. 3. What storage media to use for backup? This is normally decided by the average volume of data that is backed up every time the backup procedure is executed. Small backups are normally taken on floppy disks, whereas larger backups are taken on magnetic tapes. Very large backups may also be taken on disk packs. 4. Who takes backup? This normally depends on the usage environment of computer systems. Foi example, in case PCs it is the PC user who is normally responsible for taking the backup of his/her. PC. In case of large computer centres, where a system is shared by many users, backup is normally the responsibility of the system administrators of the system. Often backup is an automated procedure in large computer centres, in which case it is automatically executed at the defined periodicity and the data on hard disk is automatically backed up on an on-line backup device such as a tape library. In this case, backup is the responsibility of the system itself. Where to store the backup media with the backed up data? We just now saw that the backup device along with the media may be an on-line device such as a tape library. In this case, the backup media is actually stored very near to the computer system. This is not acceptable if we need to protect the data against natural calamities. Hence off-line storage of backup data is always preferred to its online storage. Moreover, to protect the data against natural calamities, it is suggest to store the backup media.with the backed up data in a building that is away from the building where the computer system whose data is backed up is located. RAM Disk We saw that hard disks are used as on-line secondary storage devices in most computer systems of today. However, the data access speed of a hard disk is much slower than that of a semiconductor RAM (recall that RAM chips are used for main memory). This is because hard disks involve mechanical movements, whereas semiconductor RAMs are pure electronic devices. To make applications execute faster, in some computer systems a block of semiconductor RAM chips (other than the RAM chips used for the main memory) is used as a simulated disk. That is, an application to be executed is first loaded from the hard disk into the simulated disk and then only its execution is started. The application executes faster than normal in this case because the method avoids most of the delay that might be caused by frequent calls to the hard disk for data loading during execution. The RAM chips used as a simulated disk are often referred to as a RAM disk, or a silicon disk, or a pseudo disk. These RAM chips may be placed on an add-on circuit board that fits inside the main computer box, or they may be housed in a separate cabinet. In either case, these extra RAM chips look just like a disk drive to the control programs used in the computer's operating system. Note that RAM disk is not really a disk at all. A RAM disk greatly helps in enhancing the speed of the computer system because the CPU can now access the desired data almost instantly from the RAM disk instead of
waiting for few milliseconds to obtain the same data from the hard disk. However, RAM chips are volatile and the data stored in them are lost when power is switched off. Thus, programs and data files produced during a processing session must be stored on a permanent storage device, such as a hard disk, before the power is switched off. Digital Versatile Disk (DVD) DVD is a standard format for the distribution and interchange of digital content in both computer-based and consumer electronic products. The format is designed to store various types of large multimedia data on a new and improved version of the CD (Compact Disk) which works on optical disk technology. That is, both DVD-ROM and CD-ROM store data using pits embossed in the disk surface. However, these pits are about AXA times as dense on the DVD-ROM as on the CD-ROM and can store about seven times more data per side. The greater density is due to a more efficient data modulation scheme and error correction method that packs more data bits into every bit. Physically, both DVD and CD are identical in both diameter and total thickness. Currently available DVDs have 4.7 GB capacity per side. They provide about 135 minutes of video playback at an average data rate of little less than 5 Megabits/second. The DVD standard has two layers of specifications known as physical layer and logical layer. The physical layer specifications apply to the physical media itself, whereas the logical layer specifications are for supporting all types of applications. The physical layer specification defines the following types of physical media: 1. DVD-ROM. It is used for mass distribution of pre-recorded software programs and multimedia. 2. DVD-RAM. It is used for general read-and-write applications in the PC or consumer electronics domains. 3. DVD-R. It is -used low-cost, write-once recordable media. 4. DVD-RW. It is a rewritable version of DVD-R. On the other hand, the logical layer specification defines recording formats for video and audio data for mass distribution of pre-recorded motion pictures and music. The key features of DVD-video and DVD-audio specifications are as follows: 1. DVD-video. It offers many more features than other pre-recorded movie products. It beats the original 12-inch laser disk products in convenience and low price. It is far superior to pre-recorded tapes in image quality and sound because it has multichannel surround sound. In addition, it has multiple selectable language soundtracks and subtitles, voice-over commentaries by the movie's cast or director, and convenient random access to special features. 2. DVD-audio. It offers multiple choices of sampling rate and numbers of bits per sample. For super-high fidelity, the music producer might opt for two channels with 24bit sampling at a rate of 192 KHz. (Contrast the basic CD standard with its two channels and 16-bit sampling at 44 KHz.) The specification also supports up to six channels of multichannel surround sound with 24-bit sampling at a 48 KHz rate. Besides the
improved audio features, the DVD-audio specification goes beyond CD by making provision for visual menus, still images, and video to accompany the audio program. Storage Hierarchy We have learnt about the various types of storages used with computer systems. We also saw that different types of storages have different capacity, speed of access, and cost per bit stored. As a single type of storage is not superior in speed of access, capacity, and cost, most computer systems make use of a hierarchy of storage technologies. A typical storage hierarchy pyramid is shown in Figure 8.28. It includes cache memory, main memory, secondary storage and mass storage. As we move up the pyramid, we encounter storage elements that have faster access time, higher cost per bit stored, and less capacity. A larger storage capacity, lower cost per bit stored, and slower access time are the results of moving down the pyramid. Thus, cache memory generally has the fastest access time, the smallest storage capacity, and the highest cost per bit stored. The primary storage (main memory) falls next in the storage hierarchy list. On-line, directaccess secondary storage devices such as magnetic hard disks make up the level of hierarchy just below main memory. Off-line, direct-access and sequential-access secondary storage devices such as magnetic tape, floppy disk, zip disk, WORM disk, etc. fall next in the storage hierarchy. Mass storage devices, often referred to as archival storage, are at the bottom of the storage hierarchy. They are cost-effective for the storage of very large quantities of data when fast access time is not necessary. Notice that the use of a hierarchy of storage technologies shown in Figure 8.28, is a cost effective way of designing computer systems with very large storage capacities. Smaller capacity, faster access time, and higher cost per bit stored Cache memory Main memory On-line, direct-access and sequential-access secondary storage device such as hard disk Off-line, direct-access and sequential-access secondary storage devices such as magnetic tape, floppy disk, zip disk, WORM disk, etc. Larger capacity, slower access time, and lower cost per bit stored Mass storage devices such as tape library, CD juke box, etc. Points to Remember 1. The primary storage of a computer system has limited capacity and is volatile. Hence, additional memory, called auxiliary memory or secondary storage, is used with most computer systems. 2. The secondary storage of a computer system is non-volatile and has low cost per bit stored, but it generally has an operating speed far slower than that of the primary storage. 3. A sequential-access storage device is one in which the arrival at the location desired may be preceded by sequencing through other locations, so that access time varies
according to location. On the other hand, a random-access storage device is one in which any location in the device may be selected at random, access to the information stored is direct, and approximately equal access time is required for each location. 4. Magnetic tape is the most popular sequential-access storage device. It consists of a plastic ribbon, usually V2 inch or lA inch wide that is coated with a magnetizable recording material. The tape ribbon is stored in reels or a small cartridge or cassette. 5. The tape of a magnetic tape storage is divided into vertical columns called frames and horizontal rows called channels or tracks. A character is recorded per frame. 6. The storage capacity of a magnetic tape is a multiple of its data recording density and length. 7. A magnetic tape drive is used for storage and retrieval of data stored on magnetic tape medium. 8. A tape drive is connected to and controlled by a tape controller, which interprets the commands for operating the tane drive.200 I Foundations of Computing 9. The four commonly used types of magnetic tapes are i^-inch tape reel, V2-inch tape cartridge, Vi-inch streamer tape and 4-mm digital audio tape. 10. Magnetic disk is the most popular direct-access storage device. It consists of a thin, circular plate made of metal or plastic that is coated on both sides with a magnetizable recording material. The disk is stored in a specially design^ protective envelope or cartridge, or several of them may be stacked together in a sealed, contamination-free container. 11. The surface of a magnetic disk is divided into a number of invisible concentric circles called tracks. Each track is further subdivided into sectors. A sector is the smallest unit of data access from a disk. In case of disk units having multiple disk plates, a set of corresponding tracks on all the recording surfaces together form a cylinder. 12. The storage capacity of a disk system is a multiple of its number of recording surfaces, number of tracks per surface, number of sectors per track and number of bytes per sector. 13. The data access time for a disk depends on seek time, latency and transfer rate. The time required to position the read/write head over the desired track is called seek time. The time required to spin the desired sector under the head is called latency. Transfer rate refers to the rate at which data are read from or written to the disk. 14. A magnetic disk drive contains all the mechanical, electrical and electronic components for holding one or more disks and for reading or writing of information on to it. 15. A disk drive is connected to and controlled by a disk controller, which interprets the commands for operating the disk drive. 16. All types of magnetic disks may be broadly classified into two types - floppy disks and hard disks. Floppy disks are made of flexible plastic whereas hard disks are made of rigid metal. Depending on the type of packaging, hard disks are further classified into zip disks, disk packs and Winchester disks. 17. An optical-disk storage system consists of a rotating disk, which is coated with a thin metal or some other material that is highly reflective. Laser beam technology is used for recording/reading of data on the disk.
18. The storage capacity of an optical disk is a multiple of its number of sectors and number of bytes per sector. 19. An optical disk drive contains all the mechanical, electrical and electronic components for holding an optical disk and for reading or writing of information on to it. 20. The two most popular types of optical disks in use today are CD-ROM and WORM disks. CD-ROM disk is a read-only storage medium whereas WORM disk is a writeonce, read-many storage medium. 21. Magneto-optical disks integrate optical and magnetic disk technologies to enable rewritable storage with laser-beam technology. 22. Mass storage devices use multiple units of a storage media to create a single secondary storage device. 23. A disk array (also known as a RAID) is a mass storage device that uses a set of hard disks and hard disk drives with a controller mounted in a single box. All the disks of a disk array form a single large storage unit. RAID systems provide enhanced storage capacity, enhanced performance and enhanced reliability. 24. An automated tape library is a mass storage device that uses a set of magnetic tapes and tape drives with a controller mounted in a single unit. 25. A CD-ROM jukebox is a mass storage device that uses a set of CD-ROM disks and CD-ROM drives with a controller mounted in a single unit. 26. In order that devices manufactured by independent vendors can be used with different computer manufacturers, it is important that the device controllers follow some drive interfacing standard. The four commonly used drive interface standards are Integrated Drive Electronics (IDE), Enhanced Small Device Interface (ESDI), Intelligent Peripheral Interface-level 3 (IPI-3) and' Small Computer System Interface (SCSI). 27. Data backup is the process of creating a copy of the data from storage backup device so that if by accident the data on the on-line it can be retrieved from the backup device and stored back on the on-line storage device to a secondary storage device gets corrupted or gets lost, line storage device. 28. The two basic types of backup are - full backup an incremental backup. In full backup, all the data on the online secondary storage device is copied on the backup device at the time of backup, whereas in incremental backup only the newly created files or files that have been changed since the last backup are copied. 29. To make applications execute faster, in some computer systems a block of semiconductor RAM chips (other than the RAM chips used for the main memory) is used as a simulated disk. These RAM chips are often referred to as a RAM disk, or a silicon disk, or a pseudo disk. 30. DVD (Digital Versitile Disk) is a standard format for the distribution and interchange of digital content in both computer-based and consumer electronic products. The format is designed to store various types of large multimedia data on a new and improved version of the CD which works on optical disk technology. 31. As a single type of storage is not superior in speed of access, capacity and cost, most computer systems make use of a hierarchy of storage technologies known as storage hierarchy. A typical storage hierarchy pyramid includes cache memory, main memory, secondary storage and mass storage.
Questions 1. What is a secondary storage? How does it differ from a primary storage? 2. What are the main limitations of the primary storage of a computer system? 3. Why is secondary storage used in most computer systems? 4. Name some of the popular secondary storage devices used in today's computer systems. 5. Give the broad classification of the popular secondary storage devices used in today's computer systems. 6. What is a sequential-access storage device? Give examples of a few applications for which such a storage device is suitable. 7. What is a random-access storage device? Give examples of a few applications for which such a storage device is suitable. 8. Distinguish between a sequential access, a direct access, and a random access storage device. Give one example of each. 9. "The storage approach selected for a particular application is determined by the way the data are organized and processed." Discuss. 10. Differentiate between on-line and off-line storage of data. Name a storage device suitable for each type of data storage. 11. Explain how information js recorded on a magnetic tape. 12. What is a parity bit? How is it used for detecting errors? 13. What is a magnetic tape drive? Describe the main components and the basic operation mechanism of a half-inch magnetic tape reel drive. 14. Define "data recording density" and "data transfer rate" for a magnetic tape drive. What is the data transfer rate for a magnetic tape system for which the tape density is 800 BPI and the tape speed is 200 inches per second? 15. A magnetic tape is a continuous medium in which data are recorded serially and there is no addressing mechanism. In this situation, explain how different data records stored on a magnetic tape are identified. 16. In the context of magnetic tape storage, what is an inter-record gap? Why is it needed? 17. In the context of magnetic tape storage, what is an inter-block gap? Why is it needed? 18. In the context of magnetic tape, what is blocking? Explain how blocking helps in faster tape I/O operation. 19. In the context of magnetic tape storage, what is blocking factor? How is blocking factor generally decided' 20. The tape usage increases as the blocking factor increases. Give an example to illustrate this. 21. The tape usage increases as the blocking factor increases. This means that the blocking factor should be i large as possible. What is the problem in making the blocking factor too large? 22. In the context of magnetic tape storage, explain the role of the following: (a) File header label (d) End of tape marker (b) File trailer label (e) Beginning of tape marker (c) Tape header label
23. What is a magnetic tape controller? List some of the commands that a magnetic tape controller typical supports. 24. List out the main advantages of half-inch tape cartridges over half-inch tape reels. 25. List out the main differences in the way data are organized and accessed in case of half-inch tape reels ai half-inch tape cartridges. 26. List out the main differences between half-inch tape and quarter-inch streamer tape. Which of the two ma better utilization of tape storage and why? 27. What is the QIC standard? How is it useful? 28. Differentiate between serial and parallel representation of data recording on magnetic tapes. Give one example of each to illustrate the two types of data recording. 29. Explain how a very high data recording density is obtained in case of digital audio tape. 30. List out the main advantages of magnetic tapes as a secondary storage device. 31. List out the main limitations of magnetic tapes as a secondary storage device. 32. List out the main uses of magnetic tapes. Write short notes on: (a) Secondary storage (b) Magnetic tape drive (c) Magnetic tape reel (d) Magnetic tape cartridge (e) Streamer tape (f) Digital audio tape (g) Sequential-access storage device (h) Direct-access storage device (i) Uses of magnetic tapes (j) Advantages and limitations of magnetic tapes 34. What is a magnetic disk? Explain how data are stored and organized on a magnetic disk. 35. In the context of magnetic disk storage, define the following terms and give the relationship among them (if any): (a) Track (d) Cylinder (b) Sector (e) Disk address (c) Cluster 36. Describe the two commonly used methods of partitioning a disk surface into sectors and list out their relative advantages and disadvantages. 37. What will be the storage capacity of a double-sided disk that has 400 tracks, 16 sectors per track and 512 bytes per sector? 38. List out the relative advantages and disadvantages of sector-based and cluster-based disk organizations. 39. The sector size of a disk should not be too large or too small. Explain why. 40. What is a disk pack? Explain how data are organized and accessed from a disk pack. 41. What is a cylinder? A disk pack has 16 surfaces and 200 tracks on each surface. How many cylinders are there in this disk pack? 42. Explain how cylinder-based organization leads to faster access of data in a disk pack. 43. What factors determine the storage capacity of disks?
44. A disk pack consists of 6 disk plates. Each plate has 400 tracks and there are 50 sectors per track. If 512 bytes can be stored per sector, then calculate the total number of bytes that can be stored in this pack. 45. Disks having larger diameter can have more tracks and larger storage capacity. In spite of this fact, why are disks with very large diameters not used for designing large capacity magnetic disk storage systems? 46. Describe the structure of the two commonly used types of access arms assembly in disk systems. List their relative advantages and disadvantages. 47. What is a head crash in case of a disk system? How does it affect a computer system? 48. Define the following terms in context of a disk storage: (a) Access time (c) Latency (b) Seek time (d) Transfer rate 49. For every read/write access to a disk, is seek time constant or variable? If it is constant, what is its value? If it is variable, what are its minimum and maximum values? 50. For every read/write access to a disk, is the value of latency constant or variable? If it is constant, what is its value? If it is variable, what are its maximum and minimum values? 51. Explain how do disk systems with multiple read/write heads per surface help in reducing seek time as compared to those with single read/write head per surface. 52. What is transfer rate of a disk system? What are the two main parameters on which it depends? 53. The rotational speed of a disk system having a single recording surface is 300 rpm. It has 80 sectors/track and 512 bytes/sector. What is the transfer rate of this disk system? 54. Explain access time, seek time and latency for disk storage. What is the access time of a disk system whose average seek time is 20 milliseconds and average latency is 7 milliseconds? 55. What is "flying-head?" How does it help in increasing the life of a disk storage? 56. Differentiate between direct and random access storage devices. In which category does a magnetic disk fall? Give reasons for your answer. 57. What is disk formatting? Why is it needed? 58. Differentiate between a formatted and an unformatted disk. How is the formatting of an unformatted disl done? 59. What is a File Allocation Table (FAT) in context of a disk storage? Why is it needed? How is it created? 60. When are two computer systems said to have compatible disk drives? How is this compatibility featun useful? 61. What is a magnetic disk drive? Differentiate between disk drives with interchangeable magnetic disks am disk drives with fixed magnetic disks. 62. List out the relative advantages and limitations of disk drives with interchangeable magnetic disks and dis drives with fixed magnetic disks. 63. What is a disk controller? List out the typical commands supported by a disk controller along with th parameters to be specified with each command. 64. List out the various types of magnetic disks commonly used in today's computer systems. Give some typical uses of each type of magnetic disk.
65. List out the key differences between floppy disks and hard disks. Which one is more durable and why? 66. What is a floppy disk? Describe the basic principles of operation of a floppy disk drive. 67. List out the main differences between a 5!4-inch diskette and a 3 '/2-inch diskette. Which one is more durable and why? 68. How will you write protect a (a) 5'/i-inch diskette (b) 3'/2-inch diskette 69. List out some of the precautionary measures that one should take for the proper usage and handling of floppy disks. 70. Although floppy disks are non-volatile storage media, data stored on them can still be lost or damaged if they are not handled properly. Discuss. 71. What is "byte rot" problem? How can you take care of this problem? 72. What is a hard disk? Name three different types of hard disks. Give a typical usage of each type. 73. What is a zip disk? List out the similarities and differences between a zip disk and a floppy disk. 74. What is a zip drive? Differentiate between portable and fixed types of zip drive. 75. What is a disk pack? List out the similarities and differences between a disk pack and a Winchester disk. 76. What is a Winchester disk? Why is it so called? 77. For the same number of disk platters of the same size, Winchester disks can manage to have larger storage capacity than disk packs. Explain why. 78. A disk storage has four platters. How many recording surfaces are there in this disk storage if it is (a) a Winchester disk (b) a disk pack? 79. Hard disks outperform floppy disks in both data storage capacity and data access speed. Explain why. 80. Give two reasons why hard disks have more storage capacity than floppy disks. 81. Give two reasons why hard disks have better data access speed than floppy disks. 82. List out the advantages and limitations of magnetic disks as a secondary storage device. 83. List out the main advantages of magnetic disks as compared to magnetic tapes as a secondary storage device. 84. List out the typical uses of magnetic disks. 85. Write short notes on: (a) Disk formatting (b) Disk pack (c) Floppy disk (d) Hard disk (e) Zip disk (f) Winchester disk (g) Magnetic disk drive (h) Uses of magnetic disks
(i) Advantages and limitations of magnetic disks (j) Proper usage and handling of floppy disks 86. What is an optical disk? How are data recorded/read from an optical disk? 87. Differentiate between the ways data are organized on a magnetic disk and an optical disk. Which data organization leads to faster random access time and why? 88. It is said that an optical disk drive requires a more complicated drive mechanism than a magnetic disk drive. Explain why. 89. What are the main parameters on which the storage capacity of an optical disk depends? Write the equation for calculating the storage capacity of an optical disk based on these parameters. 90. A 5%-inch optical disk has 3,00,000 sectors, each of 2352 bytes. What is its total storage capacity? 91. Why are mechanical read/write access arms not needed in case of optical disk drives? 92. Why are optical disk drives slower (have larger access time) than magnetic disk drives? 93. List out the main differences between a magnetic disk drive and an optical disk drive. 94. What are the two commonly used types of optical disks? What is the basic difference between the two? 95. Differentiate between a CD-ROM and a WORM disk. Give some typical uses of each. 96. What is a CD-ROM? Why is it so called? 97. What is "mastering" of CD-ROM disks? How is it done? 98. What is an "infopedia"? Give some typical examples of it. 99. What is the ISO 9660 standard for? How is it useful? 100. What is a WORM disk? What limitation of a CD-ROM disk does it overcome? What limitation it still has? 101. What is meant by "data archiving"? Why is it done? Name some secondary storage media that are used for data archiving. 102. List out the main advantages and limitations of optical disks as a secondary storage device. 103. List out the reasons why optical disks are a better storage medium for data archiving as compared to magnetic tapes or magnetic disks. 104. List out the typical uses of optical disks. 105. Write short notes on: (a) Optical disk drive (b) CD-ROM (c) WORM disk (d) Uses of optical disks (e) Advantages and limitations of optical disks 106. What is a magneto-optical disk? What limitation of an optical disk does it overcome? 107. Explain how data are recorded and read from a magneto-optical disk. 108. List out some of the typical uses of magneto-optical disks. 109. What is the basic principle of having larger capacity in mass storage devices?
110. What are the three commonly used types of mass storage devices? What are the basic storage media components used in each type? 111. What is a disk array? How does it provide enhanced storage capacity and enhanced performance? 112. What is a RAID? Why is it so called? Explain how does it provide enhanced reliability. 113. What is mirroring? Explain how mirroring leads to enhanced reliability and faster read operations in case of a RAID. 114. What is striping? Explain how striping leads to enhanced reliability and faster read operations in case of a RAID. 115. What is a RAID level? Name some of the commonly used RAID levels and the techniques used byeach for enhanced reliability. 116. What is an automated tape library? What are its main components? How are these components used together to provide a mass storage device? List out some typical uses of an automated tape library. 117. What is a CD-ROM jukebox? What are its main components? How are these components used together to provide a mass storage device? List out some typical uses of a CD-ROM jukebox. 118. List out the two basic differences between an automated tape library and a CDROM jukebox. 119. Write short notes on: (a) Disk array (b) RAID (c) CD-ROM jukebox (d) Automated tape library 120. What is a drive interface standard? How is it useful? Name two commonly used drive interface standards. 121. Write short notes on: (a) Integrated Drive Electronics (IDE) interface (b) Enhanced Small Device Interface (ESDI) (c) Intelligent Peripheral Interface-level 3 (IPI-3) (d) Small Computer System Interface (SCSI) 122. Explain how SCSI device interface standard is different from other device interface standards like IDE and ESDI. 123. What is (a) a SCSI controller, (b) a SCSI bus and (c) a SCSI device? 124. Draw a diagram to illustrate how SCSI devices are connected to and communicate with a computer system. 125. What is data backup? Why is it needed? 126. Name some secondary storage devices that are commonly used for data backup. 127. Name four different ways in which data on an on-line storage device may get damaged or lost. 128. Name the two basic types of backup. Which one will normally take more time and storage and why? 129. Differentiate between full backup and incremental backup. List out their relative advantages and disadvantages.
130. In case of incremental backup, only the newly created files or files that have been changed since the last backup are copied on the backup device from the on-line secondary storage device at the time of backup. Explain how a backup utility keeps track of the newly created and modified files since the last backup. 131. What is a backup policy? Why is it needed? What types of issues does it deal with? 132. What is meant by periodicity of backup? How is it normally decided? 133. What is automatic backup? How is it different from a manual backup? 134. How can critical data be protected against natural calamities? 135. Write short notes on: (a) Incremental backup (b) Full backup (c) Backup policy — 136. What is a RAM disk? How does it help in faster execution of an application? 137. What is a pseudo disk? How does it help in enhancing the speed of a computer system? 138. List out the relative advantages and disadvantages of a RAM disk and a magnetic disk. 139. What is DVD? Why is it so called? 140. Differentiate between a DVD-ROM and a CD-ROM. 141. Differentiate among DVD-ROM, DVD-RAM, DVD-R and DVD-RW. Give a typical use of each. 142. What is DVD-video? List out its key features. 143. What is DVD-audio? List out its key features. 144. What is a storage hierarchy? Draw a typical storage hierarchy pyramid. 145. Explain how the use of a storage hierarchy helps in building very large storage capacities for computer systems in a cost effective way. 146. Write short notes on: (a) RAM disk (b) Data backup (c) Storage hierarchy 147. Write short notes on: (a) DVD-video (b) DVD-audio 148. Write short notes on: (a) Magnetic tape (b) Magnetic disk (d) Mass storage devices (e) Drive interface standards (f) Digital Versatile Disk (DVD) (c) Optical disk (d) Magneto-optical disk 149. Write the full form of the following abbreviations: (a) DAT (b) IRG (c) IBG (d) BOT marker
(e) EOT marker (f) DDS format (g) QIC standard (h) FAT (i) DSDD diskette (j) DSHD diskette (k) CD-ROM (1) WORM disk (m) CD-R drive (n) IDE interface (o) ESDI (P) IPI-3 (q) SCSI (r) DVD (s) DVD-ROM (t) DVD-RAM (u) DVD-R (v) DVD-RW (w) RAID (x) VTOC Chapter 9 Input-Output Devices We have already seen in Chapter 2 that data and instructions must enter the computer system before any computation can be performed and the results of computation must be supplied to the outside world to allow the users to take decisions and necessary actions. Thus, a computer system can be useful only when it is able to communicate with its external environment (its users). As shown in Figure 9.1, the input-output devices (abbreviated I/O devices) provide the means of communication between the computer and the outer world. They are also known as peripheral devices because they surround the CPU and the memory of a computer system. Input devices are used to enter data from the outside world into primary storage and output devices supply the results of processing from the primary storage to the users. A wide variety of I/O devices are now available. For a particular application, one type may be more desirable than another. There are some devices, which are used for both the input and the output functions. The goal of this chapter is to familiarize the readers with the various types of I/O devices available for computer systems. It is important to note that even the fastest of the I/O devices is very slow when compared to the speed of primary storage and CPU. The main reason for this is that the speed of I/O devices in most cases depends upon mechanical movement and the potential for improvement of such devices is limited. It has been difficult to produce I/O devices that can match the processor and storage speeds, and there is a constant demand for faster and faster I/O devices. INPUT DEVICES An input device is an electromechanical device that accepts data from the outside world and translates them into a form the computer can interpret. A wide range of input devices is available today that can be broadly classified into the following categories: 1. Keyboard devices 2. Point-and-draw devices 3. Data scanning devices 4. Digitizer 5. Electronic cards based devices 6. Voice recognition devices
7. 8.
Vision based devices Offline data entry devices
The various types of input devices along with their typical applications are described below. Keyboard Devices Keyboard devices are the most commonly used input devices today. They allow data entry into a computer system by pressing a set of keys (labeled buttons), which are neatly mounted on a keyboard connected to the computer system. Keyboard devices can be broadly classified into two types - general-purpose keyboards and special-purpose keyboards. They are described below. General-purpose Keyboards General-purpose keyboards are standard keyboards used with most computer systems. They are called general-purpose because they have enough keys to make them useful for any type of application. That is, they are designed to meet the data entry requirements of a very wide range of computer users. The most popular general-purpose keyboard used today is the 101-keys QWERTY keyboard. It was introduced and promoted by IBM for use with IBM PCs and IBMcompatible PCs (personal computers). It is so called because it has altogether 101 keys out of which the keys for alphanumeric characters follow the commonly used QWERTY layout. The QWERTY layout has been adopted from the standard typewriter keyboard layout. The positioning of the character keys in this layout ensures faster typing speed. In this layout, the alphabetic keys are arranged so that the upper-left row of letters begins with the six letters Q, W, E, R, T, and Y (see Figure 9.2). Designers of other keyboards claim that their boards are easier to learn and use than the QWERTY keyboard. One such example is the Dvorak keyboard. It has not been widely accepted, however, because most people have already learnt and got used to the QWERTY keyboard layout. As shown in Figure 9.2, the keys of a QWERTY keyboard are grouped into the following functional groups to make the keyboard easier and more comfortable to use: 1. Alphanumeric Keys. The alphanumeric keys contain the keys for the letters of English alphabet, the numbers 0 to 9, and several special characters like ?, /, $, &, @, etc. As already mentioned, the arrangement of these keys follows the QWERTY layout. 2. Numeric Keypad. The numeric keypad is a set of keys that looks like an adding machine with its ten digits and mathematical operators (+, -, » and /). It is usually located on the right side of the keyboard. Notice from Figure 9.2 that the ten digits and the mathematical operators keys, which form the numeric keypad are also present in the set of alphanumeric keys. This makes the numeric keypad redundant. However the numeric keypad is still provided because it permits rapid entry of numeric data. As it is redundant, numeric keypad is usually not provided on portable PCs (laptops) due to space limitations.
3. Arrow Keys. Arrow keys are a set of four keys - up (T), down (I), left (<-) and right (->). They are also called cursor-control or cursor-movement keys because they are mainly used to move the cursor at a desired location on the terminal screen connected to the keyboard. A cursor, which usually appears as an underscore (_), a vertical line (|), or a rectangle (■) on the terminal screen, always indicates the location of the next keyed-in character on the screen. The arrow keys allow you to move the cursor up and down, usually a line at a time, and left and right, usually a character at a time. For many software packages, the arrow keys are also used to view parts of a document or worksheet that extend past the size of the terminal screen. This is known as scrolling. The up and down keys are used for vertical scrolling and the left and right keys are used for horizontal scrolling of the contents displayed on the terminal screen. 4. Special Function Keys. The keyboard also has a set of special function keys, each one of which is labeled with the type of function it performs. Some of the most important keys that belong to this set are: - Enter. Is used to send keyed-in data or a selected command to the memory. Shift. Is used to enter a capital letter or one of the special characters above the numbers. This is achieved by pressing the Shift and the corresponding character keys simultaneously. Ctrl. Is used in conjunction with other keys to expand the functionality of the keyboard. For example, just as the Shift key is used with an alphabet key to enter the alphabet in capital letter, Ctrl key is also used with another key to give a key new meaning. The exact functionality in this case depends on the software being used. Alt. Similar to Ctrl key, this key is also used in conjunction with other keys to expand the functionality of the keyboard. - Backspace. Is used to move the cursor one position to the left and delete the character in that position. - Tab. Is used to advance the cursor to the next user-defined tab position, which is the location where the user wants to enter the next data. - Caps Lock. Is used to toggle (alternate/switch) the capital letter lock feature on and off. When this feature is "on", all letters entered from the keyboard are entered as capital letters. - Num Lock. Is used to toggle the number lock feature on and off. When this feature is "on", numbers are entered when you press the numbered keys on the numeric keypad. - Home. Is used to move the cursor to the beginning of the work area, which may be the beginning of the line, screen or document depending on the software being used. - End. Is used to move the cursor to the end of a work area. - Delete. Is used to delete the character at the cursor position. - Page up. Is used to display the previous page of the document being currently viewed on the terminal screen. The page size in this case is equal to the size of the screen. - Page down. Is used to display the next page of the document being currently viewed on the terminal screen. Page up and Page down keys are also used to position the cursor at the previous and next record when using database software.
- Insert. Is used to toggle between the insert and typeover modes of data entry. In insert mode, any text entered gets inserted (added) at the cursor position. On the other hand, in typeover mode, the entered characters get typed over the character at the cursor position. - Escape. Is usually used to negate the current command or menu and return to the work screen. This key may have other functions depending on the software being used. - Spacebar. Is used to enter a space at the cursor position. - Print screen. Is used to print what is currently being displayed on the screen. 5. Other Function Keys. Besides the keys mentioned above, the keyboard also has a set of user-programmable function keys that are merely labeled as Fl, F2, F3, and so on. The actual function assigned to each of these keys is different for different software packages. Hence these function keys are used in many different ways by the programs being executed. For example, a word processing program can use the Fl, F2 and F3 keys to make the selected text underlined, italic and bold respectively. These function keys are also known as soft keys because their functionality can be defined in software. Special-purpose Keyboards The data entry requirements of computer systems deployed for special-purpose applications may not be as versatile as general-purpose computer systems. Hence computer systems used for special-purpose applications can have special-purpose keyboards to enable faster data entry and rapid interaction with computer systems. For example, the point-of-sale (POS) terminals used in supermarkets and fast-food restaurants for registering sold goods and issuing a receipt to every customer can have a simpler keyboard equipped with special-function keys that enable rapid processing of sales transactions as well as alphanumeric input. For instance, instead of typing the name and price of a purchased item, say Veg. Pizza, the operator need only press the key labeled "Veg. Pizza" and the system automatically prints the name and price of the item from an internally stored database. This application requires a special-purpose keyboard for rapid data entry and faster operation. Similarly, an ATM (Automatic Teller Machine) used in banks for financial transactions by customers needs a simple keyboard with only a few keys for the purpose. There are many such applications that may have the benefit of rapid data entry and faster operation by the use of special-purpose keyboards that are custom-designed for a special type oi application. Keyboards specially designed and used for a specific type of application are called special-purpose keyboards. Other Related Concepts Some other concepts related to keyboards are discussed below. Foreign Language Keyboards Several countries, which do not use English alphabets in their local languages, have designed special keyboards for data entry in their local languages. Some of these keyboards have a large number of keys and use a coding scheme with more bits to accommodate the large number of characters used in the language. For example, the
coding scheme used for the QWERTY keyboards works with an 8-bit code, which can accommodate 2 = 25€ different characters. However, Japanese language keyboards work with a 16-bit coding scheme to accommodate all the characters used in Japanese language. Keyboard Templates We saw in the discussion above that the exact function assigned to the function keys Fl to F12 and to some of the special function keys depends on the software being used. That is, different software packages program the same keys to perform different functions. If a user uses different types of software, it becomes difficult for him/her tc remember the functions assigned to different keys for different software. Hence several software vendors distribute their software packages with keyboard templates that designate which commands are assigned to which function keys. These templates are usually made of plastic that nicely fit over the keyboard. The functionality 01 the various function keys is written on tpp of the plastic at the positions of those function keys. With this, a used need not remember the functions assigned to different keys by different software packages. Software Simulated Keys We often come across the situation of entering a non-standard character such as "a" or "±", or a phrase or name such as "Prentice-Hall of India" again and again while preparing a particular document. To allow faster entry of data, a user may reassign a keyboard key to produce them. For example, a user may write a software to reassign the character "#" to the name "Prentice-Hall of India" so that whenever he/she enters the character "#" from the keyboard, the system automatically enters "Prentice-Hall of India" in place of the character #. Such arrangements of making the key produce something different from what it is normally supposed to produce is known as software simulation of the key. Autorepeat Feature Almost all keyboards are designed with the auto repeat feature. According to this feature, if a key remains pressed for the auto repeat initial delay period (usually of the order of 300 milliseconds),, then the character or the functionality of the key starts auto-repeating. For example, if the key with the character "a" is kept pressed, then the characters "aaaaa....." will start appearing on the terminal screen till it is kept pressed. This feature facilitates faster data entry when the same character is to be entered multiple times or when multiple lines or pages have to be skipped. Keyboard Buffer A keyboard buffer is a small memory inside either the terminal or the keyboard in which each keystroke is first saved before being transferred to the computer's main memory for processing. The keyboard buffer usually has capacity to save from a few tens to a few hundreds of keystrokes. Storing of keystrokes in keyboard buffer instead of sending them directly to the main memory makes the user interaction with the system much more
efficient and smooth. This is because when the keystroke is made, the processor may be busy doing something else and may not be in a position to immediately transfer the keyboard input to the main memory. In this case, if there was no keyboard buffer, the keyboard would have got locked after every keystroke, not allowing the user to enter anything till the keyboard input was transferred to the main memory and the lock was released. This would have caused severe problem and delay in data entry especially for data entry operators with fast typing speeds. The keyboard buffer helps in overcoming this problem by allowing the user to enter keyboard inputs at their own convenient speed. The keyboard inputs are stored in the keyboard buffer and the processor keeps transferring them to the main memory whenever it gets time to attend to the keyboard device. Point-and-Draw Devices Interaction with computers was initially restricted mainly to text mode. However, it was soon realized that interacting with computers in text-mode is cumbersome and timeconsuming. Hence a new type of interface, called graphical user interface (GUI) was devised for interacting with computers. As shown in Figure 9.3, a GUI provides to the user a screen full of graphic icons (small images on the screen) or menus and allows the user to make a rapid selection from the displayed icons or menus to give instructions to the computer. With such an user interface, the basic requirement is to have an input device that can be used to rapidly point to and select a particular graphic icon or menu item from the multiple options displayed on the screen. The keyboard was found to be very inconvenient and unsuitable for this requirement. Hence research efforts to find a suitable input device to meet this requirement gave birth to several input devices like mouse, track ball, joystick, light pen and touch screen. Later it was realized that many of these devices like mouse and light pen could also be very effectively used to create graphic elements on the screen, such as lines, curves, and freehand shapes. With this new ability, these devices came to be known as point-and-draw devices. These input devices have made computers a much easily usable tool and have also established the computer as a versatile tool for graphic designers. Some of the most commonly used point-anddraw devices are described below. Mouse As of today, the mouse is the most popular point-and-draw device. It has become a musthave input device on personal computers and workstations that have a GUI-based video display terminal. As shown in Figure 9.4, the mouse is a small hand-held device, which can comfortably fit in a user's palm. It rolls on a small bearing and has one or more buttons on the top. When a user rolls the mouse across a flat surface, such as on top of the table on which the computer is placed, the graphics cursor moves on the screen of the video display terminal in the direction of the mouse's movement. The graphics cursor, which is used to point and draw, is displayed on the screen as a variety of symbols such as an arrow ( fl ), a wrist with a pointing finger (m~), etc. Depending on the application, the text and graphics cursor may be displayed on the screen at the same time.
Since all movements of the mouse are reproduced by the graphics cursor on the screen, you can move the graphics cursor at a menu item or an icon by moving the mouse. For example, move the mouse away from you to move the cursor up on the screen, move the mouse towards you to move the cursor down on the screen, move the mouse right to move the cursor right on the screen, and move the mouse left to move the cursor left on the screen. In this way, the mouse can be used for quick positioning of the graphics cursor on the desired menu item or an icon. When positioned at a menu item or an icon, the graphics cursor is said to point to that menu item or icon. With a click of the mouse's button, the system can then be notified of this choice. Note that notifying the system of a particular choice out of the various options provided by the software is much easier with a mouse than pressing various key combinations. With the proper software, a mouse can also be used to draw pictures on the screen and edit text. The following five simple techniques are used to carry out the various types of operations possible with a mouse: 1. Point. To point means to move the mouse suitably to move the graphics cursor on top of an icon or a menu item. 2. Click. To click means to press and release the left mouse button once. It allows selection of the menu item represented by the icon, which is currently being pointed to by the graphics cursor. The function of the right button varies between software packages, but often it is used to call up a menu of options. 3. Double-click. To double-click means to press and release the left mouse button twice in rapid succession. It allows execution of the program represented by the icon, which is currently being pointed to by the graphics cursor. The function of doubleclicking the right mouse button varies between software packages. 4. Simultaneous-click. To simultaneous-click means to press and release both the left and right mouse buttons simultaneously. It is used by some software packages to give added functionality to the mouse. The actual functionality for simultaneous-click varies from one software package to another. 5. Drag. To drag an item means to position the graphics cursor over the item, then depress the button and hold it down as the graphics cursor is moved on the screen by moving the mouse. You may drag an icon to change its location on the screen. When using a graphics software program, you drag the graphics cursor to draw graphics objects (such as lines, circles, curves, freehand drawings, etc.) on the screen. When using a word processing program, you select a block of text for deleting it or for changing its style by dragging the graphics cursor from the beginning to the end of the block. The mouse reports the movement and button click events in a sequential stream as they happen. These are coded as multibyte records. The first byte is the event type, and the following bytes are the particulars of the event (the change in the vertical and horizontal position of the mouse for mouse movement events, which button and the type of click for mouse click events). However, like a keyboard, a mouse also uses an interrupt mechanism to report the mouse operations to the computer system. That is, it does not directly send a message to the program that is currently executing.. Rather, it stores the sequential stream of events in a special memory buffer and sends an interrupt request to the CPU. The currently executing program regularly checks to see whether a mouse event
has occurred. On sensing an interrupt signal, »the program reads the sequential stream of events from the special memory buffer and then reacts appropriately. Mouse devices are of many types and there are many ways to categorize them. Some of the commonly used ways to categorize mouse devices are discussed below. 1. Mechanical and Optical Mouse. A mechanical mouse has a ball inside it that partially projects out through an opening in its base. The ball rolls due to surface friction when the mouse is moved on a flat surface. On two sides of the ball at a 90-degree angle from each other are two small wheels that spin to match the speed of the ball. Each wheel of the mouse is connected to a sensor. As the mouse ball rolls when the mouse is moved, the sensors detect how much each wheel spins and send this informationjo the computer in the form of changes to the current position of the graphics cursor on the screen. An optical mouse has no mechanical parts like the ball and wheels. Instead it has a builtin photo-detector. Such a mouse comes with a special pad with gridlines printed on it. For moving the graphics cursor on the screen, the mouse has to be moved on the special pad. The photo-detector senses each horizontal and vertical line on the pad as the mouse moves on it, then it sends this information to the computer in the form of changes to the current position of the graphics cursor on the screen. 2. One, Two and Three Buttons Mouse. A mouse can have one, two or three buttons. As many programs are designed to use only a single button, so a one-button mouse is good enough for them. Some programs are designed to operate with a two-button or three-button mouse. If the system has lesser number of buttons, then such programs can still be made to run on the system by providing an alternative way to indicate a click of these buttons - such as by pressing the shift key on the keyboard while the mouse button is clicked. With a mouse having multiple buttons, the leftmost button is the main button that allows for most mouse operations because most people are right-handed and use their right index finger to click the main button. However, another button can always be configured as the main button. For example, the right-most button can be configured as the main button for left-handed users. 3. Serial and Bus Mouse. A serial mouse is one, which is designed to plug into a serial port (a built-in socket in the computer's back panel to which external devices can be connected). Such a mouse does not require a special electronic card for connecting the mouse to the computer system. On the other hand, a bus mouse is one that requires a special electronic card that provides a special port just for the mouse for connecting the mouse to the computer system. The special electronic card is fitted into one of the expansion slots inside the computer. Thus a bus mouse occupies one of the expansion slots whereas a serial mouse does not need any expansion slot. 4. Wired and Cordless Mouse. A wired mouse is connected to the computer system with a small cord (wire). On the other hand, a cordless mouse is not connected to the computer system. It communicates with the computer system with the help of a special controller that operates by transmitting a low-intensity radio or infrared signal. Although a cordless mouse is more expensive than a wired mouse, it allows easier movements of the mouse.
Trackball A trackball is a pointing device that is similar to a mechanical mouse. As shown in Figure 9.5, the ball that is placed in the base of a mechanical mouse is placed on the top along with the buttons in case of a trackball. To move the graphics cursor around the screen, the ball is rolled with the fingers. Because the whole device is not moved for moving the graphics cursor, a trackball requires less space than a mouse for operation. Since it need not be moved for moving the graphics cursor, it is often attached to or built into the keyboard. Trackballs built into the keyboard are commonly used in laptop (notebook) computers because a mouse is not practical for laptop users in a small space. Because the ball is fixed on the top, some people find it helpful to think of a trackball as an upsidedown mouse. A trackball comes in various shapes and forms with the same functionality. The three commonly used shapes are a ball, a button, and a square. In case of a ball, the ball is rolled with the help of fingers to move the graphics cursor. In case of a button, the button is pushed with a finger in the desired direction of the graphics cursor movement. In case of a square plastic, the finger is placed on top of it and moved in the desired direction of the graphics cursor movement. Joystick A joystick is a pointing device that works on the same principle as a trackball. To make the movement of the spherical ball easier, the spherical ball, which moves in a socket, has a stick mounted on it (see Figure 9.6). Instead of using the fingers in case of a trackball, the user of a joystick moves the spherical ball with the help of the stick with his/her hand. The stick can be moved forward or backward, left or right, to move and position the graphics cursor at the desired position. Potentiometers are used to sense the movements. On most joysticks, a button on the top is provided to select the option, which is currently pointed to by the cursor. The button is clicked to make this selection. Typical uses of a joystick include video games, flight simulators, training simulators, and for controlling industrial robots. Electronic Pen Another point-and-draw device is an electronic pen. In a pen-based system, you hold the pen in your hand and directly point with it on the screen to select menu items or icons; or directly draw graphics on the screen with it; or write with it on a special pad for direct input of the written information to the system. Electronic pens are of the following types: 1. Light Pen. It uses a photoelectric (light sensitive) cell and an optical lens mounted in a pen-shaped case. The photoelectric cell and lens assembly is designed in such a way that it focuses onto it any light in its field of view. Hence with this mechanism the pen can be used to indicate screen position to the computer. You operate the pen by moving its tip over the screen surface. As the light pen is moved over the screen surface, it detects the light being emitted from a limited field of view of the monitor's display. This electric
response is transmitted to a processor, which can identify the menu item or icon that is triggering the photocell. In this manner, the graphics cursor automatically locks on to the position of the pen and tracks its movement over the screen. The case of the pen has a finger-operated button, which is used to select an option from the screen by clicking it. In addition to being used for pointing and selecting options from a set of options displayed on the screen, light pens are frequently used for computer-aided design (CAD) applications. For example, the user can directly draw on the screen with a light pen. By using the pen along with a keyboard, the CAD application can allow users to select different colors and line thicknesses, reduce or enlarge drawings, and add or erase graphics objects such as lines, circles, rectangles, etc. 2. Writing Pen with Pad. This type of electronic pen comes with a special type of writing pad. The pen is used to write on the pad whatever information the user wants to input to the computer system. Although this input device may appear to be an easy way to input text into the computer for word processing, as of today it is not used for inputting large amounts of text. The main reason for this is that the technology to correctly decipher the handwriting of the users is not yet matured and successful. However, this does not mean that the writing pen and pad is not so useful input device. It is a very useful input device for signatures and messages that are stored and transferred as a graphic image without the need for the system to decipher it as is done in case of a FAX message. The image of the signature or message can be stored by the computer for retrieval by the same user at a later date, or for transfer to another user's system who can easily make sense out of the message when it is displayed on his/her screen. This input device is commonly used in Personal Digital Assistants {PDAs), which are pocket-sized computers having calendar, calculator, and notepad facility. The notepad facility is in the form of a writing pen with a pad and allows the user to make notes of information to be stored and retrieved at a later point of time. Note that the notepad allows the user to input not only textual information but also freehand drawings. Pocket computers with writing pen and pad input device are also used by goods delivery organizations in which the person making deliveries can have recipients sign their name on the computer with a pen. The image of the signature is captured and stored like a photograph without any need for the computer to decipher it. Touch Screen Touch screen is the most simple, intuitive, and easiest to learn of all input devices. A touch screen enables the users to choose from available options by simply touching with their finger the desired icon or menu item displayed on the computer's screen. Computers with touch screen facility use optical sensors in, or near, the computer's screen that can detect the touch of a finger on the screen. The sensors communicate the position of touch to the computer, which interprets the input made by the user. Another technology used by touch screen monitors is the pressure sensitive monitors. In this case, a flat pad is located beneatn the base of the computer's screen, which has sensors to measure the monitor's weight at many points. When a user touches the screen, the changing weights and forces transferred down to the sensors allow the device to detect
the location of the touch. Note that this type of touch screen monitors may sometimes require the user to apply a little pressure with his/her finger at the point of touch to successfully transmit the desired input to the system. Touch screens are very useful for computer systems that need to have a simple, intuitive interface with the intended users so that the users can use the system without any formal training. Hence they are the most preferred human-computer interface devices used in information kiosks. An information kiosk is an unattended system that is used to store information of public interest and allows common people to access the stored information as per their requirement. For example, information kiosks may be located 1. At an airport or a railway station to provide information to arriving passengers about hotels, restaurants, tourists spots, etc. in the city. 2. In large department stores and supermarkets to provide information to the customers about the locations of the various types of goods being sold, locations of the rest rooms, locations of other attractions in the shopping complex, the items on sale on a particular day and their locations, etc. 3. In large museums or zoos to guide the visitors to the locations of various attractions and facilities as well as cautioning them against things that they are not supposed to do while inside. At all these public places, a system with a touch screen interface is most intuitive and desirable for use by common public. Touch screen interface is also used in automatic teller machines by banks and credit card companies. They are also appropriate for use in environments where dirt or weather conditions would prohibit the use of keyboards and pointing devices. Data Scanning Devices In order to reduce the possibility of human error in data entry, the trend in data entry has been towards decreasing the number of transcription (copying data by typing, keying or writing) steps. This is accomplished by entering the data directly from the source document. Data scanning devices are input devices used for direct data entry into the computer system from source documents. Some of the data scanning devices are also capable of recognizing marks or characters. Data scanning devices typically have the following characteristics: 1. They eliminate some of the duplication of human effort required to get data into the computer. Human beings do not have to key the data. 2. The reduction in human intervention improves data accuracy and can increase the timeliness of the information processed. 3. Since scanners are direct data entry devices, they demand high quality of input documents. Documents that are poorly typed or have strikeovers or erasures are normally rejected. 4. With these devices, form design and ink specification may become more critical than is the case when people key the data from the forms. Data scanning devices are of many types. Some of the commonly used ones are described below.
Image Scanner An image scanner is an input device that translates paper documents into an electronic format that can be stored in a computer. The input documents may be typed text, pictures, graphics, or even handwritten material. This input device has been found to be very useful in preserving paper documents in electronic form. The copy of a document stored in a computer in this manner will never deteriorate in quality or become yellow with age and can be displayed or printed whenever desired. If the computer in which the scanned document is stored has the right kind of software (called image-processing software), the stored images can be altered and manipulated in interesting ways. Image scanners are also called optical scanners because they use optical technology for converting an image into electronic form. That is, they convert an image into electronic form by shining light onto the image and sensing the intensity of the reflection at every point of the image. This conversion process can be best understood with the help of an example. Suppose there is a document with the word "BIT" written on it. This document has to be scanned and stored in the computer. The scanner treats the document as if it is drawn on a graph sheet as shown in Figure 9.7(a). An intersection of a horizontal and a vertical line of the graph is called a grid point. If there is a dark spot at a grid point, it is represented by a 1, otherwise it is represented by a 0. Figure 9.7(b) shows the representation of the image of Figure 9.7(a). This representation is called the bit map representation of an image. Each bit in the representation of the image is called & pixel (a picture element). The finer is the granularity of grid points, the better the scanner is because it can convert a given image with greater accuracy. Scanners having 300 to 600 grid points per inch are common. Moreover, we saw in Figure 9.7 that each grid point is encoded as just 1 bit. More bits are used to represent each grid point in high quality black-and-white scanners and color scanners. For example, a good quality black-and-white scanner may use 4 bits per grid point to encode different shades from black to gray to white. Similarly, a color scanner may use 24 bits per grid point to represent a maximum of 256 different shades of color. Image scanners come in various shapes and sizes. The two commonly used types based on this classification are flat-bed and hand-held scanners. They are briefly described below. 1. Flat-bed scanner. As shown in Figure 9.8, a flat-bed scanner is like a copy machine consisting of a box having a glass plate off its top and a lid that covers the glass plate. The document to be scanned is placed upside down on the glass plate. The light source is situated below the glass plate and moves horizontally from left to right when activated. After scanning one line, the light beam moves up a little and scans the next line. The process is repeated for all the lines. It takes about 20 seconds to scan a document of size 21 cm x 28 cm.
2. Hand-held scanner. As shown in Figure 9.9, a hand-held scanner has a set of light emitting diodes encased in a small case that can be conveniently held in hand during operation. To scan a document, the scanner is slowly dragged from one end of the document to its other end with its light on. The scanner has to be dragged very steadily and carefully over the document, otherwise the conversion of the document into its equivalent bit map will not be correct. Due to this reason, hand-held scanners are used only in cases where high accuracy is not needed. They are also used when the volume of documents to be scanned is low. They are also much cheaper as compared to flat-bed scanners. Optical Character Recognition (OCR) Device We saw in the description above that image scanners may be used to input a wide variety of documents including pictures, graphics, typed text, or handwritten text. When they are used for inputting text documents (typed c handwritten), they have the following two limitations: 1. Since the input document is stored as an image instead of text, it is not possible to do any word processing of the document (the computer cannot interpret the stored document as letters, number and special characters). 2. The storage required for storing the document as an image is several times more than that required fc storing the same document as a text. For example, a page of printed text having 2000 characters can be stored as 2000 bytes using the ASCII representation. A bit map image representation of the same document will require 10 to 15 times more storage depending on the resolution (granularity) of gri points. The OCR technology is used to overcome these limitations. In this case, the scanner is equipped with a character recognition software (called OCR software) that converts the bit map images of characters to equivalent ASC codes. That is, the scanner first creates the bit map image of the document and then the OCR software translate the array of grid points into ASCII text that the computer can interpret as letters, numbers and special characters. To translate bitmaps into text, the OCR software examines the bitmap of each character and compares with the set of characters the machine has been programmed to recognize. Whichever character pattern it matches, or near matches, is considered to be the character read. If a scanned character does not match satisfactorily with any c the already stored character patterns, it is rejected. OCR software is extremely complex, because it is difficult to make a computer recognize an unlimited number (typefaces and fonts. Hence these software are designed to recognize texts that are written using standard type. fonts (called OCR fonts). Two such standard fonts are OCR-A (American standard) and OCR-B (Europea standard). OCR-A and OCR-B fonts are shown in Figure 9.10. and Figure 9.11. respectively. Note that if tl document contains italics or bold face letters, or fonts other than that for which the OCR software has bee designed, the OCR software will not work effectively.
So far, OCR devices work reliably only for printed text. However, attempts are also being made these days to use OCR devices for recognizing handwritten characters. In this case, special care has to be taken to ensure that in the handwritten document, characters are of standard size, lines making up the characters are connected, and no stylish loops etc. are used. There is yet no successful OCR software to fully convert handwritten text into ASCII. The main reason behind this is that everyone's handwriting is different and the parameters for how individual characters are formed are very broad. In fact, programming a computer to recognize handwriting is so tough that it may be several years before computers can read most handwriting accurately. The difficulty in teaching computers to recognize handwritten documents can be understood by the fact that the handwriting of many individuals is almost illegible to other individuals, especially if it is examined letter by letter. In such cases we take advantage of the fact that when we read a handwritten document, we have a good idea of its context (what the topic is). Hence we can often read a handwritten document even if we can decipher only a few letters in each word. But teaching a computer to perform the same task is really difficult. Optical Mark Reader (OMR) These scanners are capable of recognizing a pre-specified type of mark made by pencil or pen. For example, many students might have appeared in objective type tests where they had to mark their answers to questions on a special pre-printed test scoring sheet by darkening a small square, circular or oval shaped box by a pencil to indicate their correct choice out of the various given options (see Figure 9.12). These answer sheets are directly fed to a computer for grading with the use of an optical mark reader. The actual technique used by an OMR device for recognition of marks involves focusing a light on the page being scanned and detecting the reflected light pattern from the marks. Pencil marks made with a soft lead pencil (high graphite content) reflect the light enabling the reader to determine which responses are marked. Note that OMR devices require that the pre-printed forms use good quality paper and the boxes to be marked on them are accurately aligned for correct interpretation of input data. Furthermore, the pre-printed forms should not be folded so that the marks can be read correctly. The use of OMR is not limited only to the grading of objective type tests. In fact, any input data that is of a choice or selection nature can be recorded for OMR input. Thus OMR is very useful for grading tests with objective type questions in which a large number of students appear and for researchers who need to tabulate responses to large surveys where responses can be restricted to one or more out of a few possibilities. Bar-Code Reader Data coded in the form of small lines (known as bars) are known as bar codes. Bar codes represent alphanumeric data by a combination of adjacent vertical lines (bars) by varying
their width and the spacing between them. They are particularly used for unique identification of all types of goods, books, postal packages, badges, tags, etc. A bar-code reader is a device used for reading (decoding) bar-coded data. It may be a hand-held scanner, or may be embedded in a stationary scanner. It scans a bar-code image and converts it into an alphanumeric value, which is then fed to the computer to which the bar-code reader is connected, just as though the alphanumeric value had been typed on a keyboard. A bar-code reader uses a laser-beam scanning technology. The laser beam is stroked across the pattern of bars of a bar code. Different bar codes having different patterns of bars reflect the beam in different ways, which is sensed by a light-sensitive detector. Reflected light patterns are converted into electrical pulses and then transmitted to recognition logic circuits which convert it into an alphanumeric value. Just as there are a variety of internal bit encoding systems, there are a variety of barcoding systems. The most widely known bar-coding system is the Universal Product Code (UPC), which now appears on almost all retail packages in USA. The UPC, originally used for supermarket items, is now being printed on other consumer goods also. The UPC bar-code patterns are decoded as 10 digits. The first 5 of these digits identify the manufacturer or supplier of the product and the next 5 digits identify a specific product of the manufacturer (see Figure 9.13). Bar code readers are commonly found in supermarkets and department stores. When a customer picks up a product for purchasing and/brings it to the cash counter for payment, the sales person at the counter uses a barcode reader to scan the bar code printed on the product. The bar-code reader converts the bar code into an alphanumeric value and feeds it to the computer that looks up the price of the product, possibly updates inventory and sales records, and uses the price and description information to prepare a purchase bill for the customer. Postal delivery services also use bar codes to identify and track each packet. For example, Federal Express of USA uses a unique bar-coding system and their employees can usually tell a customer within a matter of minutes the current location of any packet. Magnetic-Ink Character Recognition (MICR) MICR is similar to OCR and is used by the banking industry for faster processing of the large volume of cheques being handled every day by this industry. Banks that employ MICR technology use a special type of cheque. The bank's identification code (name, branch, etc.), account number, and the cheque number are pre-printed (encoded) using characters from a special character set on all these cheques with a special ink that contains magnetizable particles of iron oxide before the cheques are given to the customers for use. A sample cheque that employs MICR technology is shown in Figure 9.14.
When a customer presents a filled-in cheque at a bank, a bank employee manually enters (keys in) the amount written on the cheque in the lower right corner of the cheque using an MICR inscriber, which prints the amount with the magnetic ink. The date of the transaction is automatically recorded for all cheques processed that day. This cheque is then processed using an MICR reader-sorter, which can recognize magnetic ink characters. The MICR reader-sorter reads the data on the cheques and sorts the cheques for distribution to other banks or for further processing. The most commonly used character set by MICR devices is known as E13B font that consists of the numerals 0 to 9 and four special characters as shown in Figure 9.15. Coded data in the form of these fonts are transferred from cheques to the computer by an MICR reader-sorter. As the cheques enter the reading unit, they pass through a magnetic field, which causes the particles in the ink to become magnetized. Read heads then interpret these characters by examining their shapes. The sorter is basically used to sort the cheques into different pockets according to their identification code numbers. MICR technology speeds up data input for the banking industry because cheques can be directly fed into the input device and automatically processed thereafter. Besides enabling faster processing of cheques, this technology also ensures accuracy of data entry because most of the information is pre-printed on the cheque and is directly fed to the computer. However, MICR has not been adopted by other industries because the currently supported character set has only 14 symbols (10 digits and 4 special characters). No alphabetic characters are available. Digitizer A digitizer is an input device used for converting (digitizing) pictures, maps and drawings into digital form foi storage in computers. For example, the JC and y coordinates of points in a drawing may be stored in digital form. This enables re-creation of the drawing from the stored information whenever required as well as easy incorporation of changes in the drawing as and when required. As shown in Figure 9.16, a digitizer consists of a digitizing tablet (also known as graphics tablet) associated wit! a stylus. The digitizing tablet is a flat surface that contains hundreds cf fine copper wires forming a grid. Each copper wire receives electric pulses. The digitizing tablet can be spread over a working table and is connected to a computer. The stylus is like a pen or a lens-like cursor with a cross-hair and button. The stylus is connected to the tablet and can be pressed down at a point on the tablet to input the (x, y) coordinates of the point. When the stylus is moved on the tablet, the cursor on the computer's screen moves simultaneously to a corresponding position on the screen to provide visual feedback to the operator. This enables the operator to draw sketches directly or to input sketched drawings very easily. Inputting drawings or developing sketches using a digitizer is further simplified by the fact that poorly sketched lines, arcs and other graphical objects are automatically input as mathematically precise objects, like straight lines and smooth curves.
Digitizers are commonly used in the area of Computer Aided Design (CAD) by architects and engineers to design cars, buildings, medical devices, robots, mechanical parts, etc. They are also used in the area of Geographical Information System (GIS) for digitizing of maps available in paper form. Electronic-card Reader Electronic cards and their associated readers offer another means of direct data entry into a computer system. Electronic cards are small plastic cards having encoded data appropriate for the application for which they are used. An electronic-card reader, which is normally connected to a computer, is used to read the data encoded on an electronic card and transfer it to the computer for further processing. Electronic cards are often used by banks and issued to the customers for use in automatic teller machines (ATMs). An ATM allows a customer to deposit or withdraw cash 24 hours a day without the need to interact with a bank employee. It is an unattended machine, which may be located at any place (either inside the bank or away from the bank). It is normally connected to the bank's computer. The electronic card for use in an ATM has the customer's account number and credit limit (maximum amount that he/she can withdraw in a day) encoded on t card. An ATM has an electronic-card reader associated with it. When a customer inserts his/her card into the ATM (actually the electronic-card reader of the ATM), the reader reads the data encoded on the card and transmits it to the bank's computer, which activates the customer's account. The customer can then carry out t desired transaction by following the instructions displayed on the screen of the ATM and pressing a few keys on the customized keyboard of the ATM. Electronic cards are also used by many organizations for controlling access of various types of employees to physically secured areas. In this case, each employee is given an electronic card with his/her authorization data encoded on it. The authorization data contains information that specifies which employee is allowed access which areas. All physically secured areas of the organization have an electronic-card reader installed at entrance. When an employee wants to enter into a physically secured area, he/she inserts his/her electronic ca into the reader installed at the entrance of that area. The reader reads and checks the authorization data encoded on the card before permitting the individual to enter the area (the door is electronically linked to the reader and opens only if the reader finds that the individual is authorized to have access to the area). In such applications, the electronic-card reader is normally linked to a central computer, which maintains a record of who all entered or left secured area and at what time. In case of security breaches, this record is used to trace the problem. Depending on the manner in which data is encoded, electronic cards are of the following two types: 1. Magnetic strip cards. These cards have a magnetic strip on their back. The data to be encoded the card is stored on the magnetic strip. The magnetic strip of a card can contain much more data f unit of space than do printed characters or bar codes. Moreover, the
data stored on the magnetic strip cannot be read visually. This makes them a perfect candidate for storing confidential data. 2. Smart cards. Smart cards have a built-in microprocessor chip instead of the magnetic strip. The data to be encoded on the card is permanently stored in the memory of the microprocessor chip. The memory of a smart card microprocessor can contain much more data than the magnetic strip of a magnetic strip card. Moreover unlike a magnetic strip card, a smart card also has some processing capability because its microprocessor has its own processor and memory. This feature attributes intelligence to smart cards and makes them more suitable for a wider variety of applications than magnetic strip cards. For example, a smart card can be used as a cash card by storing a specific amount of cash in the microprocessor chip before the card is issued to a customer. As and when the customer uses the card to make purchases, the amount of purchase made at every instance is deduct from the balance amount stored in the card by inserting the card in a card read/write machine (al called electronic registers) that is available with shops who allow use of smart cards for making purchases. At any time, the customer can pay more money to the card company and get the balance amount stored in the card increased. In effect, this application allows smart cards to be used electronic money. Because smart cards can hold more information, have some processing capability, and is almost impossible to duplicate, they may soon replace magnetic strip cards. Voice Recognition Devices Voice recognition devices are input devices that allow a person to input data to a computer system by speaking it. Thus they make computers much easier to use. However as a data input device, currently voice recognition systems have limited success because correct interpretation by a machine of the large number of words in the vocabulary of a language is difficult. The major difficulty has been that people speak with different accent (pronounce differently) and intonations (with different tone or pitch of the voice), and the fact that the meaning of words can vary depending on the context in which they are used. Hence today's voice recognition systems are limited to accepting few words within a relatively small domain and can be used to enter only limited kinds and quantities of data. A computer with a voice recognition device consists of a microphone or telephone and follows the following steps to recognize and act upon a spoken input data (see Figure 9.17): 1. Speak. The person willing to input data to the computer system speaks into the microphone/telephone attached to it. 2. Convert into electrical signals. The spoken word(s) is then converted into electrical signals by breaking down each sound into its various frequencies. This step converts the spoken word(s) into its analog form. 3. Digitize. A digital-to-analog converter is then used to convert the analog form of the spoken word(s) into its digital form (in 1s and 0s) so that they can be stored and interpreted by the computer.
4. Match with pre-stored words. The digitized version of each spoken word is matched against similarly formed pre-stored words in the computer's electronic dictionary. The creation of the electronic dictionary (a database of pre-stored words) is done during a training session. That is, the system is initially operated in a training mode when the words and phrases to be stored in the electronic dictionary are spoken several times to train the system to recognize them. In this mode, the patterns for all the spoken words and phrases spoken during the training session are created and stored for future matching. 5. Perform corresponding action. When a match for the spoken word(s) is found, it is displayed on the computer's terminal and/or the appropriate action corresponding to the input is performed (for example, stop doing whatever it was performing). In some case, the system may display the word(s) and wait for operator's confirmation to prevent doing something different from what the operator intended (in case of wrong interpretation of the spoken word(s) by the computer). If no match is found for the spoken word, the speaker is asked to repeat the word. Voice recognition systems are normally classified into the following two categories: 1. Speaker-dependent. Due to the vast variation in the accent of different speakers, most voice, recognition systems of today are speaker-dependent. That is, they can recognize the speech of only a single individual or a few individuals. All these individuals must have participated in the training session to train the system to recognize their accent of the pre-stored words in the computer's dictionary. The system usually maintains a different database of pre-stored words for different individuals because the digital form of the same word may be different for different individuals. 2. Speaker-independent. Speaker-independent voice recognition systems can recognize words spoken by anyone. Based on the idea of making a speaker-dependent system for more than one individual, it is clear that speaker-independent systems will require a very large database of pre-stored words to accommodate anyone's voice pattern. To take care of this practical problem, speaker-independent systems are designed to have a very limited vocabulary. For example, the vocabulary of such a system may have the words YES, NO, ZERO, ONE, TWO, THREE, FOUR, FIVE, SIX, SEVEN, EIGHT, NINE and TEN. Speaker-independent systems are much more difficult to build than speakerdependent systems. I According to another classification mechanism, voice recognition systems are of the following two types: 1. Single word recognition systems. These systems are designed to recognize only a single spoken word, such as YES, NO, MOVE, STOP, at a time. Speaker-independent systems are mostly of this type. 2. Continuous speech recognition systems. These systems are designed to recognize spoken sentences such as MOVE TO THE NEXT BLOCK. Since recognizing continuously spoken sentences is more difficult than recognizing words, continuous speech recognition systems are speaker-dependent. Although in its infancy, voice recognition systems are already being used for a wide range of applications. Some of its typical applications are as follows:
1. For inputting data to a computer system by a person in situations where his/her hands are busy, or his/her eyes must be fixed on a measuring instrument or some other object For example, a quality-control person who must use his/her hands to detect defects can call out the identification number of the sample found defective as they are detected; persons handling baggage at airports can simply speak out the three-letter destination identifier for the luggage to be automatically routed to the appropriate conveyor system; and doctors in an operation room can request certain information about a patient while operating. 2. For authentication of a user by a computer system based on voice input. 3. For limited use of computers by individuals with physical disabilities. In addition to making input of data easier, voice recognition systems also provide tremendous freedom of movement to the operator because the operator is free to stand up and move around while inputting voice data to the system. Vision-Input System A vision-input system allows a computer to accept input just by seeing an object. The input data in this case is normally an object's shape and features in the form of an image. The idea is to simulate the capability of a human vision system in a limited sense. A computer with a vision-input device consists of a digital camera and follows the following steps to recognize a given object: 1. The digital camera is focused on the input object to take a picture of the object. 2. The digital camera creates an image of the object in digital form (in 1s and 0s) so that it can be stored and interpreted by the computer. 3. The digitized image of the object is matched against similarly formed pre-stored images in the computer's image database. 4. Depending on whether a match is found or not, the system takes the appropriate action. Vision-input devices are mainly used today in factories for designing industrial robots that are used for quality-control and assembly processes. For example, a robot used for quality-control may inspect objects and separate those, which do not meet certain qualitycontrol specifications. Another robot may be used to separate objects of different shapes in different bins. Still another robot may be used to recognize, grasp, insert, and align the various components needed to assemble some product. A simple example of a visioninput system is shown in Figure 9.18. This system separates round and square shaped objects passing on a conveyor belt in their respective bins and also separates defective objects in a reject bin. Off-line Data Entry Devices Data entry is the process of transforming input data into a form the computer system can store and interpret. Data entry may be carried out in either offline or online manner. Online data entry involves input devices, which are connected directly to the computer. On the other hand, offline data entry involves devices through which data is recorded on some media (such as floppies or tapes) and then entered into the computer later. Since data entry is done by data entry operators and the speed with which a human being can
enter data is very less as compared to the data processing speed of a computer, so when data entry, is done offline, it saves the precious computer time. As the major forms of storage media are tapes, floppies and disks, there are offline data entry devices for each of these. They are briefly described below. Key-to-Tape A key-to-tape device, also known as magnetic tape encoder, is designed to record keyed data directly on to a magnetic tape. The device has a keyboard, a display monitor and a slot to insert the tape. An operator keys in the source data using the keyboard. To facilitate data entry, a blank form with appropriate fields is displayed on the monitor's screen. The entered data is stored temporarily in a buffer (temporary storage area) by the device and displayed on the monitor for visual checking of its correctness. If an error is made in entering a data item, the operator can back-space and re-enter it. When the operator is convinced that the data entered in the form is correct and wants to store it on the tape, he/she presses a control key, which causes the data to be stored on the tape from the buffer. Magnetic tape encoders are available for recording data magnetically on reels, on cassettes, and on tape cartridges. The magnetic tape reels produced by key-to-tape systems are in a computer-compatible format for subsequent direct data input into a computer. However, data on cartridges and cassettes often are transferred to higher-speed media, such as full sized reel of magnetic tape or magnetic disk, for efficient data transfer to the computer. Key-to-Floppy The device has a keyboard, a display monitor and a slot to insert the floppy disk. Data is entered exactly in the same manner as described for the key-to-tape device. Both the key-to-tape and the key-to-floppy devices can also be used to verify the correctness of the data, which has already been stored on a tape/floppy. For this, the machine is operated in a "verify" mode and another operator keys in the same data from the source document with the tape/floppy, having the pre-stored data, inserted in the slot. In this mode, the data keyed in and stored in a buffer is compared with the data in the tape/floppy by reading the data from the tape/floppy. If4he two data do not agree, an audible and visual signal alerts the operator. The operator verifies the correctness of the field in error with the source document and re-enters the correct data if necessary which replaces the incorrect data on the tape/floppy. Key-to-Disk Key-to-disk devices are used as data recording stations in systems where data from different points have to be recorded for processing at one point. Typically, from 8 to 64 keyboard stations are linked to a small special-purpose computer to record source data on a disk. Each station has a keyboard and a display monitor. Information entered via the
keyboard is displayed on the monitor's screen to allow a visual check. The small specialpurpose computer is required to control the input from the various stations, enabling the data to be held temporarily for verification and editing before allocation to the disk storage. Thus, these data entry systems are able to correct data before storing it on a magnetic disk and before its entry into the main computer system. OUTPUT DEVICES An output device is an electromechanical device that accepts data from a computer and translates them into a form suitable for use by the outside world (the users). A wide range of output devices is available today that can be broadly classified into the following categories: 1. Monitors 2. Printers 3. Plotters 4. Computer output microfilm 5. Screen image projector 6. Voice response systems The various types of output devices along with their typical applications are described below. Output devices generate computer output, which can be broadly classified into the following two types: 1. Soft-copy output. A soft-copy output is an output, which is not produced on a paper or some material that can be touched and carried for being shown to others. They are temporary in nature and vanish after use. For example, output displayed on a terminal screen or spoken out by a voice response system are soft-copy output. 2. Hard-copy output. A hard-copy output is an output, which is produced on a paper or some material that can be touched and carried for being shown to others. They are permanent in nature and can be kept in paper files or can be looked at a later time when the person is not using the computer. For example, output produced by printers or plotters on paper are hard-copy output. Monitors Monitors are by far the most popular output devices used today for producing soft-copy output. They display the generated output on a television like screen (see Figure 9.19). A monitor is usually associated with a keyboard and together they form a video display terminal (VDT). A VDT (often referred to as just terminal) is the most popular input/output (I/O) device used with today's computers: That is, it serves as both an input and an output device. The keyboard is used for input tu the computer and the monitor is used to display the output from the computer. The name "terminal" comes from the fact that a terminal is at the terminal, or end, point of a communication path. The two basic types of monitors used today are cathode-ray-tube (CRT) and flat-panel. The CRT monitors look much like a television and are used with non-portable computer
systems. On the other hand, the flat-panel monitors are thinner and lighter and are commonly used with portable computer systems like notebook computers. How a CRT Monitor Displays an Output? A CRT monitor is a glass tube with a screen at one end and an electron gun at the other end. Its major components are shown in Figure 9.20. The electron beam generator has an oxide-coated metal called cathode and a heating filament. The cathode emits a large number of electrons when heated by the filament. In front of the electron beam generator is a control grid that is applied with an appropriate positive voltage to allow the exact number of electrons, out of the generated ones, to pass through it for striking at the monitor's screen. Larger is the positive voltage applied to the control grid, larger will be the number of electrons that will be allowed to strike the screen, and larger will be the beam's intensity. If the grid voltage is negative, then no electrons will be allowed to pass the grid and the beam is said to be blanked. The electrons, which are allowed to pass through the control grid are accelerated and focused by an assembly unit consisting of accelerating electrodes and a focusing electron lens. The accelerating electrodes are applied with a positive voltage to accelerate the electrons and the focusing electron lens consists of a cylindrical metal enclosure with a hole through which the electrons pass in the form of an electron beam. The electron beam generator, the control grid, and the accelerating electrodes with focusing electron lens together form the electron gun of the CRT monitor. To focus the electron beam generated by the electron gun at a desired point on the screen, the beam is passed through a pair of vertical deflection plates and a pair of horizontal deflection plates. Appropriate positive voltages are applied to one of the vertical and one of the horizontal plates to achieve this. For example, the beam can be moved up/down by applying a suitable positive voltage to the top/bottom plate of the vertical plates pair. Similarly, the beam can be moved left/right by applying a suitable positive voltage to the left/right plate of the horizontal plates pair. The focused and accelerated electron beam can thus be made to hit at any point on the screen. The backside of the monitor's screen is coated with phosphor, a chemical that glows for a short time when it is hit by the electron beam. The screen's phosphor coating is organized into a grid of dots called pixels (a contraction of picture elements). Since the phosphor glows only for a short time (few microseconds), an image drawn on the screen would be visible only for a few microseconds and then fade away, unless the image is repeatedly drawn. This may appear to be an inconvenience, but it is also an advantage since there is no need to erase anything. Instead, we just need to draw new images and the old ones automatically fade away. In fact, the electron gun does not draw an image directly by focusing and shooting on the pixels forming the image. Instead, it traverses the screen in a fixed pattern as shown in Figure 9.21. That is, it systematically aims at every pixel on the screen, starting at the top-left corner and scanning to the right horizontally, then dropping down a small
distance and scanning another line. Each horizontal line of pixels is called a scan line. There are a fixed number of pixels on every scan line, and a pixel can be written or not in each position by turning the beam on and off. When the beam gets to the right end of a scan line, it is turned off and moved to the starting point of the next scan line. This is called the horizontal trace. After scanning all the scan lines, when the beam gets to the right end of the bottom-most scan line, it is turned off and moved to the starting point of the top-most scan line. This is called the vertical trace. In one pass of scanning the entire screen, if the electron beam travels all the scan lines sequentially from top to bottom, it is called a. progressive scan (see Figure 9.21 (a)). On the other hand, if the electron beam travels only 9.21 (b)). In case of interlaced scan since the monitor scans every other line instead of each line in order, so it has to transmit only half as much data in a single pass, but takes two passes to draw the entire screen. As interlaced scan may cause flicker of images drawn on the screen, it is used in lowerquality monitors. High quality monitors use progressive scan. How to Compare CRT Monitors CRT Monitors are normally compared based on the following attributes: 1. Screen Size. Display screens of monitors vary in size from 5 to 25 inches (diagonal dimension). Monitors with 17 inch to 21 inch screen size are common today with personal computers and other desktop computers. With users increasingly viewing multiple windows, video clips, animated objects, and complex graphics, monitors with larger screens are preferred. However, the larger the screen the costlier is the monitor. 2. Resolution. Resolution refers to the number of points to which electron beam can be directed (pixels) on the screen. As the pixels are arranged in scan lines, the resolution of a monitor depends on its total number of scan lines and total number of pixels per scan line (note that each scan line has same number of pixels). The total number of scan lines is called a monitor's vertical resolution and the total number of pixels per scan line is called its horizontal resolution. The overall resolution of a monitor is expressed as a multiple of its horizontal and vertical resolutions. A low-resolution monitor has a resolution of about 320 x 200. Today, most monitors have a resolution of about 800 * 600. High-end monitors can have resolutions of 1024 x 768, 1280 x 1024 or even more. The higher the resolution of a monitor, the clearer will be its displayed images. Very high-resolution monitors project extremely clear images that look almost like photographs. 3. Refresh Rate. We saw that when a pixel on the phosphor-coated screen is hit by the electron beam, it glows only for a short time (few microseconds) and then fades away. Hence to make a displayed image appear flicker-free, it should be drawn again and again. The number of times it is to be drawn depends upon the persistence of human vision and persistence time of the phosphor used. Normally an image has to be redrawn 30 to 60 times a second to make it appear flicker-free. We saw that the electron beam scans the entire screen for drawing images on the screen. Hence to make the displayed contents appear flicker-free, the electron beam should scan the entire screen 30 to 60 times a second. The actual number of times that the electron beam scans the entire screen per second is called the refresh rate of the monitor. The higher the refresh rate, the better will
be the display quality of a monitor because it will be less strenuous for the eyes to continuously view the screen. Today, most monitors operate at 60 Hz. That is, they refresh the screen 60 times per second. Better quality monitors operate at 70 to 90 Hz, refreshing the screen 70 to 90 times per second. 4. Color. Based on this attribute, monitors may be either monochrome or color. A monochrome monitor can display images only in a single color, usually white, red, green, blue, or amber. On the other hand, a color monitor can display multi-colored images. Although a monochrome monitor can display only a single color, it is usually capable of displaying multiple shades (increased and reduced color intensities) of the same color. The term gray scales is used to refer to the number of shades of a color that a monitor can display. Monochrome monitors are rapidly becoming a thing of the past, as most applications today require color monitors. A color monitor functions just like a monochrome monitor. To add color, it uses three electron beams instead of one, and each pixel on the screen is made up of three tiny red, green, and blue phosphors arranged in the shape of a triangle. When the three electron beams are combined and focused on a pixel, the phosphors of that pixel light up to form a tiny spot of white light. Different colors can be displayed by combining various intensities of the three beams. Color monitors are also called RGB monitors because they use red, green and blue as the primary additive colors for generating multiple colors. Color monitors use multiple bits to represent each pixel. For example, if each pixel is represented by one byte (8 bits), then it can indicate one of 2s = 256 different colors to draw the pixels. That is, the monitor is said to support 256 different colors. There are millions of colors when we consider all the shades and mixtures of color possible, and so 256 seems like far too few colors. Today's high-end monitors usually use 24 bits to represent each pixel. Such monitors can support 224 or about 16 million different colors. Based on the resolution and number of colors supported, several standards for color monitors have evolved. Each one is implemented by installing an add-on board (commonly known as a graphics adapter board or a video card) in the main computer box and connecting it to the appropriate monitor. The four most popular of these standards are: - CGA (Color Graphics Adapter). It has a low resolution of 320 x 200 and supports up to 16 colors. - EGA (Extended Graphics Adapter). It has a resolution of 640 * 350 and supports up to 16 colors. - VGA (Video Graphics Array). It has a resolution of 640 x 480 and supports up to 256 colors. - Super VGA. It provides resolution from 800 x 600 to 1280 x 1024 and supports 256 or more colors. Today, CGA and EGA are standards of the past because all modern personal computers come with at least VGA output. A video card (graphics adaptor card) has its own memory called video RAM. The higher the resolution and the more colors a monitor supports, the larger is the video RAM required. The video RAM capacity required for different types of monitors is summarized in Figure 9.22.
Flat Panel Monitors A flat panel monitor (shown in Figure 9.23) is only about one inch thick, is light in weight, and consumes less electricity as compared to a CRT monitor. These features make a flat panel monitor most suitable for use with portable computer systems such as a laptop (notebook) computer. Note that a portable computer must be compact in size, light in weight, and should consume less power so that it can be operated for longer durations with batten at places where there is no power point, such as while traveling. These requirements render a CRT monitor unsuitable for these systems because of its large size, heavy weight, and more power requirement. The display technologies used by flat panel monitors include liquid-crystal display (LCD), electro-luminescent display (ELD), and gas-plasma display (GPD). Out of these, LCD monitors are most commonly used. As the name implies, LCD monitors use a special kind of liquid crystals to display images on the screen. The liquid crystals used in these monitors are normally transparent, but become opaque when charged with electricity. One problem with these monitors is that they are hard to read under certain lighting conditions. This is because unlike the phosphors used in CRT monitors, the liquid crystals do not glow when charged, so there is not enough contrast between the images and their background to make them clearly legible under all lighting conditions. To take care of this problem, some LCD monitors put a light source behind the screen. Although this makes the screen easier to read, the light source requires additional power, which can be a problem for portable computer systems that often run on battery. Printers Printers are the most commonly used Output devices today for producing hard-copy output. There are many ways to classify a wide variety of printers available today. The most commonly used ways are: 1. Printing technology used. According to this classification, the common types of printers are dot-matrix printers, ink-jet printers, drum printers, chain/band printers and laser printers. 2. Approach of printing. According to this classification, printers are classified as impact printers and non-impact printers. Impact printers use the approach of hammering a typeface against paper and inked ribbon to produce printed output. Nonimpact printers do not hit or impact a ribbon to print. They use thermal, electrostatic, chemical or inkjet technology to produce printed output. Impact printers can produce multiple copies by using carbon in a single print cycle, which is not possible with a non-impact printer. However, an impact printer is noisy in operation and normally uses covers to reduce the noise level. Non-impact printers are quiet in operation. 3. Speed of printing. According to this classification, printers are classified as character printers, line printers and page printers. As the names imply, a character printer prints one character at a time, a line printer prints one line at a time, and a page printer prints an entire page at a time. Accordingly, printer speeds are measured in characters-per-second
(cps) for character printers, lines-per-minute (lpm) for line printers, and pages-per-minute (dpi) for page printers. The print-speed ranges for the three types of printers are 40 to 450 cps, 1000 to 5000 lpm, and 4 to 800 ppm. 4. Quality of printing. The quality of the printed output is directly proportional to the resolution of printing which is measured in dots per inch (dpi). The dpi qualifier refers to the number of dots that can be printed per linear inch, horizontally or vertically. That is, a 300-dpi printer is capable of printing 90,000 (300 x 300) dots per square inch. According to this classification, printers are classified as near-letter-quality (NLQ), letterquality (LQ), near-typeset-quality {NTQ) and typeset-quality (TQ). Near-letter-quality printers have resolutions of about 300 dpi, letter-quality printers have resolutions of about 600 dpi, near-typeset-quality printers have resolutions of about 1200 dpi, and typesetquality printers have resolutions of about 2000 dpi. NLQ and LQ printers are used for ordinary day-to-day printing jobs, whereas NTQ and TQ printers are used for top-quality professional printing such as in publishing industries. 5. Language scripts. Since printers are output devices meant for producing hard copy output for users, they must be capable of producing output in different languages familiar to different groups of users. In fact, printers are available in almost all scripts, such as, English, Devnagari, Japanese, Arabic, etc. According to this classification, printers may be classified based on the scripts (also called fonts) supported. Note that a printer may simultaneously support multiple scripts in which case it may be used to print in multiple languages. For example, a printer may be capable of printing in both English and Devanagari. 6. Color. Printers may be capable of producing output in a single color (such as blank, red, yellow), in two or three colors (such as a document printed in black and white can have the key words printed in red), or in multiple colors. Although printers capable of printing in a single or two to three colors are suitable for producing textual documents, it is often desirable to have printers, which can print in multiple colors for producing multiple-color graphics outputs. According to this classification, printers are classified as monochrome printers and color printers. The various types of printers in use today are described below. Dot-Matrix Printers Dot-matrix printers are character printers that form characters and all kinds of images as a pattern of dots. Figure 9.24 shows how various types of characters can be formed as a pattern of dots. Based on this idea, the printing mechanism used by dot-matrix printers is illustrated in Figure 9.25. As shown in the figure, a dot-matrix printer has a print head that can move horizontally (left to right and right to left) across the paper. The print head contains an array of pins, which can be activated independent of each other to extend and strike against an inked ribbon to form a pattern of dots on the paper. To print a character, the printer activates the appropriate set of pins as the print head moves horizontally. For faster printing, many dot matrix printers are designed to print both while the print head moves from left to right and while it moves from right to left on return. A dot-matrix printer is shown in Figure 9.26.
In Figure 9.24 all the characters are formed by 5 horizontal and 7 vertical dots. Such a pattern is called a 5 x 7 matrix. It can be generated by a print head having 7 pins. However, you will notice that the print head in Figure 9.25 has 9 pins. The bottom 2 pins are used for lower-case letters that extend below the line (for example, p, g, q and y). The quality of the printed output is directly proportional to the density of the dots in the matrix. Most early dot matrix printers had 9 pins and used 5 x 7 dot matrix patterns for character formation. So the character that they printed had a maximum vertical resolution of 7 dots. Many new dot matrix printers have 18 or 24 pins, which can print near-letterquality characters that appear solid. The arrangement of pins in an 18-pin print head of a dot-matrix printer is shown in Figure 9.27. Since dot matrix printers produce printed output as patterns of dots, they can print any shape of character that a programmer can describe. This allows the printer to print many special characters, different sizes of print, and the ability to print graphics such as charts and graphs. Dot-matrix printers are impact printers because they print by hammering the pins on the inked ribbon to leave ink impressions on the paper. Hence they can be used to produce multiple copies by using carbon paper or its equivalent. However due to impact printing, dot-matrix printers are noisy as compared to non-impact printers. Since dot-matrix printers use inked ribbon to produce printed output, they are usually monochrome. However, some dot-matrix printers use a ribbon having rows of two to three colors to produce printed output in selected few colors. Figure 9.27. The arrangement of pins in an 18-pin print head of a dot-matrix printer. It permits dots to overlap to increase the244 I Foundations of Computing Dot-matrix printers are normally slow with speeds usually ranging between 30 to 600 characters per second. However they are cheap in terms of both initial cost and cost of operation. Hence they are preferred by individuals and organizations for generating printed outputs if the speed and quality of printing are not important factors. They are also used for applications that require multicopy output such as shipping forms and invoices, which rely on impact for generating multiple copies. InkJet Printers InkJet printers are character printers that form characters and all kinds of images by spraying small drops of ink on to the paper. The print head of an inkjet printer contains up to 64 tiny nozzles, which can be selectively heated up in a few micro seconds by an integrated circuit register. When the register heats up, the ink near it vapourizes and is ejected through the nozzle and makes a dot on the paper placed in front of the print head. To print a character, the printer selectively heats the appropriate set of nozzles as the print head moves horizontally. An inkjet printer is shown in Figure 9.28.
InkJet printers produce higher quality output than dot-matrix printers because they form characters by very tiny ink dots. A high-resolution inkjet printer has as many as 64 nozzles within height of 7mm, providing print resolution of around 360 dots per inch. Since inkjet printers produce printed output as patterns of tiny dots, they can print any shape of characters that a programmer can describe. This allows the printer to print many special characters, different sizes of print, and the ability to print graphics such as charts and graphs. InkJet printers are non-impact printers because they print by spraying ink on the paper. Hence they are quiet in operation. However, like other non-impact printers, inkjet printers cannot be used to produce multiple copies of a document in a single printing. InkJet printers can be both monochrome and color. A monochrome inkjet printer uses a single print head, whereas a color inkjet printer uses multiple print heads, one per color. Color inkjet printers can produce multiple color output. InkJet printers are slower than dot-matrix printers with speeds usually ranging between 40 to 300 characters per second. Typically, an inkjet printer is more expensive than a dotmatrix printer. They are preferred by individuals and organizations for generating printed outputs if the speed of printing is not an important factor. Drum Printers Drum printers are line printers that print one line at a time. The print mechanism of a drum printer is shown in Figure 9.29. It consists of a solid cylindrical drum with characters embossed (raised characters) on its surface in the form of circular bands. Each band consists of all the printing characters supported by the printer in its character set and the total number of bands is equal to the maximum number of characters (print positions) that can be printed on a line. Thus a drum printer with 132 characters per line and supporting a character set of 96 characters will have altogether 12,672 (132x96) characters embossed on its surface. In addition to the drum, the printer has a set of hammers mounted in front of the drum in such a manner that an inked ribbon and paper can be placed between the hammers and the drum. The total number of hammers is equal to the total number of bands on the drum, that is, one hammer is located opposite to each band of the drum. The drum rotates at a high speed and a character at a print position is printed by activating the appropriate hammer when the character embossed on the band at the print position passes below it. Thus the drum would have to complete one till revolution to print each line of output. This means that all characters on a line are not printed at exactly the same time, but the time required to print an entire line is so fast that it appears as if one line is printed at a time: A drum printer is shown in Figure 9.30.
The drum of a drum printer is expensive and cannot be changed often. Hence drum printers can only print a predefined set of characters in a pre-defined style that is embossed on the drum. Due to this reason, drum printers do not have the ability to print -any shape 0f characters, different sizes of print, and graphics such as charts and graphs. Drum printers are impact printers because they print by hammering the paper and the inked ribbon against the characters embossed on the drum. Hence they can be used to produce multiple copies by using carbon paper or its equivalent. However due to impact printing, drum printers are noisy in operation and often use a cover to reduce the noise level. Since drum printers use inked ribbon to produce printed output, they are usually monochrome. Typical speeds of drum printers are in the range of 300 to 2000 lines per minute. Chain/band printers are line printers that print one line at a time. The print mechanism of a chain/band printer h shown in Figure 9.31. It consists of a metallic chain/band on which all the characters of the character se supported by the printer are embossed. A standard character set may have 48, 64 or 96 characters. In order t< enhance the printing speed, the characters in the character set are embossed several times on the chain/band. For example, the chain/band of a 64 character set printer may have 4 sets of 64 characters each embossed on it. In this case, the chaia/band will have altogether 256 (64x4) characters embossed on it. In addition to the chain/band, the printer has a set of hammers mounted in front of the chain/band in such a manner that an inked ribbon and paper can be placed between the hammers and the chain/band. The total number of hammers is equal to the total number of print positions. So if there are 132 print positions, then the printer will have 132 hammers. The chain/band rotates at a high speed and a character at a print position is printed by activating the appropriate hammer when the character embossed on the chain/band passes below it. Since the character set is repeated several times on the chain/band, it is not necessary to wait for the chain/band to make a complete revolution to position the desired character in the correct print position. Unlike the drum of a drum printer, the chain/band of a chain/band printer can be easily changed. This allows the use of different fonts (styles of characters) and different scripts (languages) to be used with the same printer. However, just like drum printers, chain/band printers can only print pre-defined sets of characters that are embossed on the chain/band used with the printer. Due to this reason, chain/band printers do not have the ability to print any shape of characters, different sizes of print, and graphics such as charts and graphs. Chain/band printers are impact printers because they print by hammering the paper and the inked ribbon against the characters embossed on the chain/band. Hence they can be
used to produce multiple copies by using carbon paper or its equivalent. However due to impact printing, chain/band printers are noisy in operation and often use a cover to reduce the noise level. A chain/band printer is shown in Figure .9.32. Since chain/band printers use inked ribbon to produce printed output, they are usually monochrome. Typical speeds of chain/band printers are in the range of 400 to 3000 lines per minute. Laser Printers Laser printers are page printers that print one page at a time. The main components of a laser printer are a laser beam source, a multi-sided mirror, a photoconductive drum and toner (tiny particles of oppositely charged ink). To print a page of output the laser beam is focused on the electro statically charged drum by the spinning multi-sided mirror. The mirror focuses the laser beam on the surface of the drum in such a manner that it creates the patterns of characters/images to be printed on the page. As the drum is photoconductive, a difference in electric charge is created on those parts of the drum surface that are exposed to the laser beam. As a result, the toner, which is composed of oppositely charged ink particles, sticks to the drum in the places the laser beam has charged the drum's surface. The toner is then permanently fused on the paper with heat and pressure to generate the printed output. The drum is then rotated and cleaned with a rubber blade to remove the toner sticking to its surface to prepare the drum for the next page printing. A laser printer is shown in Figure 9.33. Laser printers produce very high quality output because they form characters by very tiny ink particles. The most common laser printers have resolutions of 600 dpi, whereas some high-end laser printers have resolutions of 120 dpi. Because of their high resolution, these printers give excellent graphics art quality. Since laser printers produce printed output as patterns generated by the laser beam, they can print any shape < characters that a programmer can describe. This allows the printer to print many special characters, different size of print, and the ability to print graphics such as charts and graphs. Laser printers are non-impact printers because they do not have hammers striking on an inked ribbon or paper. Hence they are very quiet in operation. However, like other nonimpact printers, laser printers cannot be used produce multiple copies of a document in a single printing. The majority of laser printers are monochrome, which can print multiple shades of gray. However, color laser printers, which use many different colors of toners, are also available. Color laser printers are many times more expensive than monochrome laser printers. Laser printers are faster in printing speed than other printers discussed before. Low speed laser printers can pi 4 to 12 pages per minute. Very high-speed laser printers are also
available which can print 500 to 1000 pages' minute. That is fast enough to print this entire book in about one minute. Because of their better print quality i printing speed, laser printers are more expensive than other printers. Hydra Printers Today most modern offices have the following office equipment - a printer, a fax machine, a copy machine and a scanner. There is a considerable overlap in the technologies used in these machines. The need for these mach at one place and the overlap in the technologies used in them has triggered the creation of all-inmultifunctional device that is capable of doing all four functions. Such a device is called a hydra printer. S< models of hydra printers may not perform all the four functions but may perform only two or three functions. Hydra printers are relatively new, but they may prove to be a popular office equipment of the future. Plotters We saw above that dot-matrix, inkjet and laser printers are capable of producing graphics output. However many engineering design applications, like architectural plan of a building, design of mechanical components of an aircraft or a car, etc., often require highquality, perfectly-proportioned graphic output on large sheets. The various types of printers discussed above are, not suitable for meeting this output requirement of such applications. A special type of output device called plotters is used for this purpose. Plotters are an ideal output device for architects, engineers, city planners, and others who need to routinely generate high-precision, hard-copy graphic output of widely varying sizes. The two commonly used types of plotters are drum plotter and flatbed plotter. They are briefly described below. Drum Plotter In a drum plotter, the paper on which the design has to be made is placed over a drum that can rotate in both clockwise and anti-clockwise directions to produce vertical motion. The mechanism also consists of one or more pen holders mounted perpendicular to the drum's surface. The pen(s) clamped in the holder(s) can move left to right or right to left to produce horizontal motion. The movements of the drum and the pen(s) are controlled by the graph plotting program. That is, under computer control, the drurn and the pen(s) move simultaneously to draw the designs and graphs on the sheet placed on the drum. The plotter can also annotate the designs and graphs so drawn by using the pen to draw characters of various sizes. Since each pen is program selectable, pens having ink of different colors can be mounted in different holders to produce multi-colored designs. A drum plotter is shown in Figure 9.34.
Flatbed Plotter I A flatbed plotter plots a design or graph on a sheet of paper that is spread and fixed over a rectangular flatbed table. In this type of plotter, normally the paper does not move and the pen holding mechanism is designed to provide all types of motions necessary to draw complex designs and graphs. That is, under computer control, the pen(s) move in the required manner to draw the designs and graphs on the sheet placed on the flatbed table. The plotter can also annotate the designs and graphs so drawn by using the pen to draw characters of various sizes. Here also provision is there to mount more than one pen in the pen(s) holding mechanism. Since each pen is program selectable, pens having ink of different colors can be mounted in different holders to produce multi colored designs. The plot size is restricted by the area of the bed. Some may be as small as. A4 size (8"xll" page) while some very large beds used in the design of cars, ships, aircrafts, buildings, highways, etc. can be up to 20 ft. by 50 ft. Some plotters are also designed to etch plastic or metal plates. In this case, the plastic or metal sheet is spread on the bed and the drawing pen has a sharp-edged needle. A flatbed plotter is shown in Figure 9.35. Plotters are normally very slow in motion because of excessive mechanical movement required during plotting. Hence there is a great mismatch between the speed of the CPU and the speed of a plotter. Because of this reason, in most cases, output is first transferred by the CPU onto a tape and then the plotter is activated to plot the design from the information on the tape. However, in case of a computer system dedicated to design work, the CPU may send output directly to a plotter. Computer Output Microfilm (COM) COM is an output device that allows computer output to be recorded on a sheet or roll of microfilm as microscopic filmed images. COM devices commonly use a sheet of film measuring 4"x6" for recording output information. This sheet of film is called a microfiche ("fiche" is pronounced as "fish". It is a French word that means card). COM recording process produces page images that are 48 or more times smaller than those produced by conventional printers. Hence a single microfiche can hold about 300 times as much information as a standard computer printed page. COM uses a microfilm recorder for recording output information on a microfiche, and a microfilm reader to view the information recorded on a microfiche. The microfilm recorder receives output information from the computer system which it projects pageby-page on to a CRT screen. A high-speed camera, in-built into the system, then takes a picture of the displayed information. In some systems, the recorder processes the film from the camera, and in other systems, a separate automated film developer is used to produce the filmed images of the output information. A microfilm reader that is used to view the output information recorded on a microfiche is shown in Figure 9.36. It operates on a "back projection" principle and displays a frame (a single page of information) at a time on a translucent screen whose typical size is about 8"xll". A user can read the displayed information on the screen. Most microfilm readers
also have a printer in-built into it or attached to it to allow a user to selectively print and take a hard copy of the desired pages. Most COM systems come with computer-assisted-retrieval software that allows a user to quickly search < particular document, from the thousands of pages of recorded information on the film, simply by specifying the pre-assigned index that corresponds to the desired document. Without this, users would be required to spend lots of time in manually searching through the film cartridge or card to find the desired document. As compared to a printer, a COM system has the following advantages and limitations: 1. Information recording speed is many times faster for a COM system as compared to a printer. A COM system is 20 to 25 times faster than an average line printer in this regard. 2. Microfilm output is many times more compact than a printed page output. A single microfiche can hold about 300 times as much information as a standard computer printed page. Thus the mailing and storage costs of filmed documents are much less than paper documents. 3. Microfilm output is less costly than printed output. The cost of paper needed for printing a 1000 page document is about 20 to 30 times more than the cost of the film needed to record the same information. 4. The initial investment required for installing a COM system is much more than that required for buying an ordinary printer. Hence use of a COM system is justified only in case of high volume workload. 5. Many people feel uncomfortable using a COM system because of retrieval delays and the lack of facility to write notes on the margins of the reports. A COM system provides an easy and compact way of recording and storing information, and retrieving a desired piece of information in a matter of seconds. It is ideal for use in applications where there is a large volume of information to be recorded, stored, and selectively retrieved off-line. Organizations often need to maintain records of their various types of documents, such as catalogues, bills, invoices, etc., for a number of years before destroying them. A COM system is most suitable for this purpose. Insurance companies, banks, libraries, public utilities departments, government agencies, and many other types of organizations are regular users of COM systems. Screen Image Projector Screen image projector is an output device that is used to project information from a computer on to a large screen (such as a cloth screen or a wall) so that it can be simultaneously viewed by a large group of people. This output device is very useful for making presentations to a group of people with the direct use of a computer. Before such an output device was available, the contents of a presentation were prepared using a computer, the presentation material was then printed on a printer, the printout was next reproduced on overhead projector transparency sheets using a copy machine, and finally the presentation was made using an overhead projector. Special marker pens had to be used for marking certain portions of the contents on the transparency sheets during the
presentation. A screen image projector greatly simplifies this job. It can be directly plugged to a computer system and the presenter can make a presentation to a group of the people by projecting the presentation material one after another on a large screen with the help of computer's keyboard or mouse. With this facility, the presenter can also directly point to, mark, or edit certain portions of the contents of the displayed material during the presentation to make the presentation more understandable. Additionally, a full-fledged multimedia presentation with audio, video, image and animation can be prepared and made using this facility to make the presentation more lively and interesting and also to greatly simplify the job of the presenter. Screen image projectors have become a common presentation equipment today. They are commonly used with portable notebook computers to quickly setup a modern presentation facility at any place with great ease. Like monitors, screen image projectors provide a temporary, soft-copy output. Voice Response Systems Just as a voice recognition system allows a user to talk to a computer, similarly, a voice response system enables a computer to talk to a user. A voice response system has an audio-response device that produces audio output. Obviously, the output is temporary, soft-copy output. Voice response systems are basically of two types - voice reproduction system and speech synthesizer. They are briefly described below. Voice Reproduction System A voice reproduction system produces audio output by selecting an appropriate audio output from a set of prerecorded audio responses. The set of pre-recorded audio responses may include words, phrases or sentences spoken by human beings; music or alarms generated by musical instruments; or any other type of sound. The actual analog recordings of the pre-recorded sounds are first converted into digital data and then permanently stored on the computer's disk or in its memory chip. When audio output is to be produced, the computer selects the appropriate sound from the set of pre-recorded sounds, the selected sound is converted back into analog form, which is then routed to a speaker to produce the audio output. Voice reproduction systems are very useful in a wide range of applications. Their typical uses include: 1. Audio help for guiding how to operate a system. For example, banking industry uses voice reproduction systems in automatic teller machines to provide step-by-step guidance to customers on how to transact with the bank using an ATM. 2. Automatic answering machines. For example, telephone enquiries for new telephone numbers in place of an old number, or vacancy status of a particular flight or train is often taken care of by an automatic answering machine.
3. Video games are made exciting and interesting by playing an event-based sound from a set of prerecorded sounds. 4. Talking alarm clocks. For example, every hour the clock speaks out what is the time by selecting the appropriate voice message corresponding to that hour from the set of prerecorded voice messages. Else the clock may speak "its time to wake up" at the time set for alarm. 5. Talking toys and home appliances also employ a voice reproduction system. 6. Often personal computers with audio facility are used for automated multimedia presentations during exhibitions. Speech Synthesizer A speech synthesizer converts text information into spoken sentences. To produce speech, these devices combine basic sound units called phonemes. From a given text information, sequence of words are combined into phonemes, amplified, and output through the speaker attached to the system. Speech synthesizers are still in their infancy because currently they can produce only limited unique sounds with only limited vocal inflections and phrasing. However, they are very useful in a wide range of applications. Their typical uses include: 1. For reading out text information to blind persons. For example, a recently published book may be scanned using a scanner, converted into text using OCR software, and read out to blind persons using a speech synthesizer. This will allow blind persons to know the latest information published in a book as soon as it is printed, rather than wait for the book to appear in Braille. 2. For allowing those persons who cannot speak to communicate effectively. For example, the person simply types the information and the speech synthesizer converts it into spoken words. 3. For translation systems that convert an entered text into spoken words in a selected language. For example, a foreigner coming to India may enter a text that he/she wants to communicate to an Indian and the speech synthesizer converts the entered text into spoken words of the selected Indian language. OTHER RELATED CONCEPTS Before concluding this chapter, we will have a look at a few concepts related to input/output devices. Several of these concepts are useful to understand how I/O devices are connected to and communicate with a computer system. Device Controllers We saw in Chapter 7 that all the components of a computer communicate with the computer's CPU through the system bus. That means the I/O devices need to be attached to the system bus. However I/O devices are not connected directly to the computer's system bus. Instead, they are connected to an intermediate electronic device called a device controller, which in turn is connected to the system bus. Hence a device controller
is an interface unit between an I/O device and the system bus. On one side, it knows how to communicate with the I/O device connected to it, and on the other side, it knows how to communicate with the computer's CPU and memory through the system bus. A device controller need not necessarily control a single device, rather a single device controller can usually control multiple I/O devices. It comes in the form of an electronic circuit board (also known as an electronic card) that plugs directly into the system bus, and there is a cable from the controller to each device it controls. The cables coming out of the controller are usually terminated at the back panel of the main computer box in the form of connectors known as ports. Figure 9.37 illustrates how I/O devices are connected to a computer system through device controllers. Using device controllers for connecting I/O devices to a computer system instead of connecting them directly to the system bus has the following advantages: 1. A device controller can be shared among multiple I/O devices allowing many I/O devices to be connected to. the system. 2. I/O devices can be easily upgraded or changed without any change in the computer system. 3. I/O devices of manufacturers other than the computer manufacturer can be easily plugged-in to the computer system. This provides more flexibility to the users in buying I/O devices of their choice. Device controllers may or may not have DMA (Direct Memory Access) capability. As shown in Figure 9.37, a device controller without DMA capability transfers data between the computer's memory and an I/O device via the CPU. On the other hand, a device controller with DMA capability can transfer data directly between the memory and an I/O device without the help of CPU. Obviously, device controllers with DMA capability can transfer data much faster than those without DMA capability. Due to this reason, nearly all device controllers of today have DMA capability. Serial and Parallel Ports We saw that I/O devices are connected to I/O ports of a device controller, which in turn is connected to the computer's system bus. The I/O ports are used to plug-in I/O devices to a computer system in a similar manner as electric sockets in a house are used to plug-in various types of electrical appliances when they need to be used. These I/O ports are basically of two types - serial ports and parallel ports. They are briefly described below. Serial Ports A serial port may be used to connect any device that can send or receive a stream of bytes serially, one bit at a time. A serial port is so called because it sends the data bits one at a time (that is, serially). It has only one wire for sending data and another one for receiving data. The eight bits of a byte are sent one at a time. Usually, two extra bits are sent with the byte to help in the control and timing of the communication. A chip called UART on the device controller's card converts parallel data from the system bus into serial data that
can flow through a serial cable or telephone wire. The process is illustrated in Figure 9.38. Similarly, the serial data received by the device controller from the I/O device is converted into parallel data by the UART before it is sent to the system bus. Serial ports are used for connecting slow I/O devices such as terminal, keyboard, mouse, modem, etc. The physical appearance of a typical serial port is shown in Figure 9.39. Parallel Ports A serial port limits the speed at which it can transfer data because of the fact that it can transfer only one bit at a time. Parallel ports overcome this limitation of serial ports by allowing 8 or 16 bits of data to be transferred at a time. Accordingly, a parallel port has 8 or 16 parallel wires for sending and receiving data. Hence a parallel port with 8 wires can transfer one byte (8 bits) at a time and one with 16 wires can transfer two bytes (16 bits) at a time. The process of data transfer in case of a parallel I/O port is illustrated in Figure 9.40. Notice that unlike the case of serial I/O port, the conversion of bytes into bit streams and vice-versa is not required in this case. This also leads to faster data transfer than in case of serial I/O ports. A parallel port can obviously handle a higher volume of data in the same amount of time than a serial port. Hence they are used for connecting faster I/O devices such as printer, analog-to-digital and digital-to-analog converters, etc. The physical appearance of a typical parallel port is shown in Figure 9.41. SCSI Interface A computer's system bus has a finite number of slots (called expansion slots) to which device controllers can be connected. If every I/O device comes with its own device controller card, then a computer system will have the capability to connect only a very limited number of I/O devices because it will soon run out of available expansion slots. This problem is overcome by using a SCSI interface, which is an interesting and important variation of the separate device controller idea for each device. As shown in Figure 8.27 and discussed in Chapter 8, a SCSI interface uses a generic device controller (called SCSI controller) and allows any device with an SCSI interface to be directly connected to the SCgl bus of the SCSI controller. That is, instead of connecting device controller cards into the computer's system bus via the expansion slots, SCSI extends the computer's system bus outside the computer by way of a cable, and just as you can plug one extension cord into another to lengthen the circuits, you can plug one SCSI device into another to form a daisy chain. Most current implementations of SCSI limit the number of devices that can be connected together to form a daisy chain to seven due to the bus capacity limitation. This can really save you expansion slots in your computer for connecting other devices. If none of your computer's device controller cards has a SCSI interface, you can purchase a separate SCSI interface (as the one shown in Figure 9.42) to plug into an existing parallel port to which you can connect additional SCSI devices.
Dumb, Smart and Intelligent Terminals Terminals (monitors along with keyboards) are extensively being used today as online, remote data entry devices. Depending on their local processing capability, these terminals are classified as dumb, smart and intelligent. A dumb terminal is a simple, low-cost terminal with alphanumeric display capability. It has a keyboard for input, a means of communication with the CPU of the main computer, and a screen to display the entered data and output received from the main computer. Dumb terminals need not appear dumb to the user. Because of the rapidity with which the main computer can communicate back to the terminal, many applications are not affected if no actual processing is done locally at the terminal. However, dumb terminals transmit all data directly to the main computer. This type of data is called "dirty" data because errors are transmitted along with good data. The errors are then identified and informed to the terminal operator by the CPU of the main computer and processing could be delayed until "clean" data are obtained from the operator. This type of delay, accompanied by the need to process the data over again, can be expensive. A smart terminal usually has a microprocessor (a CPU) and some internal storage in addition to I/O capabilities. It has local data editing capability and the ability to consolidate input data prior to sending them to the CPU of the main computer. This means that preliminary processing of input data can take place locally and that the data can be cleaned up at the terminal itself before it is transmitted to the main computer for processing this can lead to tremendous saving of clean data preparation time as compared to dumb terminals. However, smart terminals cannot be programmed by users. An intelligent terminal is a small computer by itself. It consists of a built-in microcomputer that can be programmed by users. These programmable terminals have extensive local data processing power. Large working files can be maintained locally and small data processing jobs can be handled by the terminal itself without the need to interact with the CPU of the main computer. They have their own on-line secondary storage devices like disk, which are used during the processing of these small jobs. Programmed error-detection tests can also be stored in the terminal for the validation of input data which can ensure the transmission of clean data to the main computer system for further processing. With the fast reduction in the cost of microcomputer components and disk storage, most terminals of today are intelligent or at least smart to help make data entry and retrieval easier for the users. X Terminals An X terminal is a graphics terminal that has a rich set of commands for the following: 1. To display characters. 2. To move the cursor position.
3. To draw graphics and images such as draw a line of desired thickness and color, draw an empty circle or a circle filled with a certain color, draw a bitmap, draw text, etc. 4. To manage areas on the screen (called windows) such as open a new window of the desired size, close an existing window, resize an existing window, etc. X terminals are very common these days. Terminal Emulators Today computers are hardly used in stand-alone mode. They are often connected to networks and communicate with other computers on the network. In this type of scenario, computers interact with each other in client-server mode and may keep changing their roles of client and server at different occasions. In such a situation, a full-fledged computer like a PC or a workstation is required to serve as a terminal for communicating with a remote computer. But a PC or a workstation is not a terminal, and does not act like a terminal. A simple way to solve this problem is to write a software program that runs on the PC or workstation and communicates with the network interface over a serial port on the PC or workstation and pretends that it is a terminal. Such a program is called a terminal emulator. Privacy Filter Privacy filters are used with monitors to prevent others sitting/standing nearby the monitor from viewing the contents displayed on it. Often users work with sensitive information displayed on their monitors in an open area and always have the fear of their co-workers around them viewing the displayed information. To overcome this problem, you can use a privacy filter, which fits over your monitor's screen, covering the displayed contents from people on either side of you. It is made of a plastic film that is inlaid with microscopic black strips. If you are sitting in front of the screen, you can see through the filter clearly. To others, the screen looks black. Economically Designed Devices Ergonomics is the science that deals with designing and arranging machines and other things in such a way that their users can use them with maximum efficiency of the. This is important for achieving maximum efficiency of the people working on these machines and protecting them from accidents or medical problems caused due to continuous usage of a machine for long hours. For example, prolonged usage of a badly designed keyboard can cause wrist problems, sometimes so serious as to require surgery. Similarly, prolonged usage of a badly designed monitor can cause eyesight problems. To help prevent these types of problems, manufactures of I/O devices are continuously working towards making ergonomically better and better devices. A few examples of efforts made in this direction are (there may be hundreds of such examples): 1. Originally, the keyboard of a computer was attached to its monitor. But today, nearly every terminal has a separate keyboard attached to the monitor with a reasonably long cable so that the user can move the keyboard to any comfortable position on the table or can even place it on his/her lap during operation.
2. Many new keyboards are designed with tilt adjustments and with hand-rest pads to make their use more comfortable for the operators. 3. Cordless mouse allows more comfortable operation of mouse device as compared to wired mouse. 4. Originally, monitors used to be fixed at one angle. But today, almost all monitors are mounted on a ball bearing and can be tilted and rotated to suit the height and angle of operation of an operator. 5. Monitors with curved screens are being replaced with those having flat screens to improve readability and reduce reflections. 6. Monochrome monitors are being replaced with color monitors because they are more comfortable to view. Many operators prefer them and report fewer instances of eye fatigue and blurred vision. 7. Newer monitors use higher resolution to prevent flicker of displayed objects, which may lead to eye problems. 8. Newer monitors also use special screen coatings for reducing glare on eyes. Points to Remember 1. The input-output devices (abbreviated I/O devices) provide the means of communication between the computer and the outer world. They are also known as peripheral devices. 2. An input device is an electromechanical device that accepts data from the outside world and translates them into a form the computer can interpret. 3. An output device is an electromechanical device that accepts data from a computer and translates them into a form suitable for use by the outside world (the users).260 m Foundations of Computing 4. Keyboard devices are the most commonly used input devices today. They allow data entry into a computer system by pressing a set of keys (labeled buttons), which are neatly mounted on a keyboard connected to the computer system. 5. A graphical user interface (GUI) provides to the user a screen full of graphic icons (small images on the screen) or menus and allows the user to make a rapid selection from the displayed icons or menus to give instructions to the computer. 6. Point-and-draw devices are input devices that can be used to rapidly point to and select a particular graphic icon or menu item from the multiple options displayed on the screen. Many of these devices can also be very effectively used to create graphic elements on the screen, such as lines, curves and freehand shapes. Some of the commonly used point-and-draw devices are mouse, trackball, joystick, electronic pen and touch screen. 7. Data scanning devices are input devices used for direct data entry into the computer system from source documents. Some of the data scanning devices are also capable of recognizing marks or characters. 8. An image scanner is an input device that translates paper documents into an electronic format that can be stored in a computer. The input documents may be typed text, pictures, graphics, or even handwritten material. Image scanners are also called optical scanners because they use optical technology for converting an image into electronic form.
9. An optical character recognition (OCR) device is a scanner equipped with a character recognition software. It is used for inputting text documents and store them in a form suitable for doing word processing of the documents. That is, the computer can interpret the stored document as letters, numbers and special characters. 10. An optical mark reader (OMR) is a scanner that is capable of recognizing a prespecified type of mark made by pencil or pen. Any input data that is of a choice or selection nature can be recorded for OMR input. 11. Bar codes represent alphanumeric data by a combination of adjacent vertical lines (called bars) by varying their width and the spacing between them. A bar-code reader is a scanner used for reading (decoding) bar-coded data. 12. Magnetic-ink character recognition (MICR) technology is used by the banking industry for faster processing of the large volume of cheques. This technology also ensures accuracy of data entry because most of the information is pre-printed on the cheque and is directly fed to the computer. 13. A digitizer is an input device used for converting (digitizing) pictures, maps and drawings into digital form for storage in computers. This enables re-creation of the drawing from the stored information whenever required as well as easy incorporation of changes in the drawing as and when required. 14. Electronic cards are small plastic cards having encoded data appropriate for the application for which they are used. An electronic-card reader, which is normally connected to a computer, is used to read the data encoded on an electronic card and transfer it to the computer for further processing. 15. Depending on the manner in which data is encoded, electronic cards are of two types - magnetic strip cards and smart cards. Magnetic strip cards store the data to be encoded on a magnetic strip attached to the back oi the card. On the other hand, smart cards have a built-in microprocessor chip and the data to be encoded or the card is permanently stored in the memory of the microprocessor chip. 16. Voice recognition devices are input devices that allow a person to input data to a computer system b) speaking to it. They are basically of two types - speaker-dependent and speaker-independent. Speaker-dependent systems can recognize the speech of only a single individual or a few individuals. On the other hand, speaker-independent systems can recognize words spoken by anyone. 17. A vision-input system allows a computer to accept input just by seeing an object. The input data in this case is normally an object's shape and features in the form of an image. 18. Online data entry involves input devices, which are connected directly to the computer, whereas offline data entry involves devices through which data is recorded on some media (such as floppies or tapes) and then entered into the computer later. Since the speed with which a human being can enter data is very less as compared to the data processing speed of a computer, when data entry is done offline, it saves the precious computer time. The commonly used offline data entry devices are key-to-tape, key-tofloppy and key-to-disk. 19. Computer output generated by output devices are of two types - soft-copy output and hard-copy output. Soft-copy outputs are temporary in nature and vanish after use. Hardcopy outputs are permanent in nature and can be kept in paper files or can be looked at a later time when the person is not using the computer.
20. Monitors are by far the most popular output device used today for producing softcopy output. They display the generated output on a television like screen. 21. A monitor and a keyboard are usually associated together to form a video display terminal (VDT), which is often referred to as just terminal. A terminal serves as both an input and an output device and is the most popular input/output (I/O) device used with today's computers. 22. Printers are the most commonly used output devices today for producing hard-copy output. Based on the printing technology used, printers are classified as dot-matrix printers, inkjet printers, drum printers, chain/band printers and laser printers. According to the approach of printing used, printers are classified as impact printers and non-impact printers. 23. Dot-matrix printers are character printers that form characters and all kinds of images as a pattern of dots. 24. InkJet printers are character printers that form characters and all kinds of images by spraying small drops of ink on to the paper. 25. Drum printers are line printers that print characters by striking a set of hammers on an inked ribbon and paper placed against a solid cylindrical drum with characters embossed on its surface in the form of circular bands. 26. Chain/band printers are line printers that print characters by striking a set of hammers on an inked ribbon and paper placed against a metallic chain/band on which all the characters of the character set supported by the printer are embossed. 27. Laser printers are page printers that produce very high quality output by forming characters and all kinds of images with very tiny ink particles. 28. Hydra printers are multifunctional devices that perform the functions of two or more of the following office equipment - printer, fax machine, copy machine and scanner. 29. Plotters are an ideal output device for architects, engineers, city planners, and others who need to routinely generate high-precision, hard-copy graphic output of widely varying sizes. The two commonly used types of plotters are drum plotter and flatbed plotter. 30. Computer Output Microfilm (COM) is an output device that allows computer output to be recorded on a sheet or roll of microfilm as microscopic filmed images. It uses a microfilm recorder for recording output information on a microfiche, and a microfilm reader to view the information recorded on a microfiche. 31. Screen image projector is an output device that is used to project information from a computer on to a large screen (such as a cloth screen or a wall) so that it can be simultaneously viewed by a large group of people. 32. Voice response systems enable computers to talk to their users. They are of two types - voice reproduction system and speech synthesizer. A voice reproduction system produces audio output by selecting an appropriate audio output from a set of pre-recorded audio responses. On the other hand, a speech synthesizer converts text information into spoken sentences. 33. A device controller is an interface unit between an I/O device and the system bus. On one side, it knows how to communicate with the I/O device connected to it, and on the other side, it knows how to communicate with the computer's CPU and memory through the system bus.
34. I/O devices are connected to I/O ports of a device controller, which in turn is connected to the computer's system bus. The I/O ports are used to plug-in I/O devices to a computer system. They are of two types -serial ports and parallel ports. 35. A serial port may be used to connect any device that can send or receive a stream of bytes serially, one bit at a time. It is used for connecting slow I/O devices such as terminal, keyboard, mouse, modem, etc. 36. A. parallel port can handle a higher volume of data than a serial port because it transfers 8 or 16 bits of data at a time. It is used for connecting faster I/O devices such as printer, analog-to-digital and digital-to-analog converters, etc. 37. SCSI interface uses a generic device controller (called SCSI controller) and allows any device with an SCSI interface to be directly connected to the SCSI bus of the SCSI controller. This helps save expansion slots in a computer for connecting other devices. 38. Depending on their local processing capability, terminals are classified as dumb, smart and intelligent. A dumb terminal simply has alphanumeric display capability. A smart terminal has local data editing capability and the ability to consolidate input data prior to sending them to the main computer. It has a microprocessor and some internal storage in addition to I/O capabilities. An intelligent terminal is a small computer by itself that has extensive local data processing power. It consists of a built-in microcomputer that can be programmed by users. 39. An X terminal is a graphics terminal that has a rich set of commands to display characters, move the cursor position, draw graphics and images, and manage multiple windows on the screen. 40. A terminal emulator is a program that runs on a PC or workstation and pretends that it is a terminal when the PC or workstation is connected to a network and communicates with other computers on the network. 41. Privacy filters are used with monitors to prevent others sitting/standing nearby the monitor from viewing the contents displayed on it. 42. Ergonomically designed I/O devices aim at achieving maximum efficiency of the users working with them by designing them in such a way that they can be used with maximum comfort and protect the users from accidents or medical problems caused due to continuous usage of a device for long hours. Questions 1. Why are I/O devices necessary for a computer system? 2. What are peripheral devices? Why are they so called? 3. Differentiate between an input device and an output device. Can a device be used as both an input device and an output device? If no, explain why. If yes, give an example of such a device. 4. Why are I/O devices very slow as compared to the speed of primary storage and CPU? 5. What is an input device? Name some of the commonly used input devices. 6. What are keyboard devices? Differentiate between general-purpose and specialpurpose keyboards. 7. Describe the organization of the keys on a QWERTY keyboard. 8. Explain the use of the following types of keys found on a keyboard: (a) Arrow keys (c) Alphanumeric keys
(b) Soft keys (d) Special function keys 9. What are soft keys on a keyboard? Why are they so called? How are they used? 10. Name some of the special function keys along with their use. 11. Give examples of two special-purpose keyboards. 12. Write short notes on: (a) Keyboard templates (b) Software simulated keys (c) Keyboard buffer (d) Foreign language keyboards (e) Numeric keypad on a keyboard (f) Autorepeat feature of keyboards 13. What is a keyboard buffer? How does it help in the operation of a keyboard? 14. What are software simulated keys? How do they help in faster data entry? 15. What are point-and-draw devices? Name some of the commonly used point-anddraw devices. 16. What is a mouse? Explain how is it used to notify the system of a particular choice out of a given set of choices on the monitor's screen. 17. What is a graphics cursor? What is its relationship with a mouse? 18. List out and describe the basic techniques used to carry out the various types of operations possible with a mouse. 19. Differentiate between a mechanical and an optical mouse. 20. Differentiate between a serial and a bus mouse. 21. Differentiate between a wired and a cordless mouse. 22. What is the minimum number of buttons required in a mouse? How does it help having more number of buttons than the minimum requirement? 23. What is a trackball? How is it different from a mouse? 24. Explain how a trackball is used to notify the system of a particular choice out of a given set of choices or the monitor's screen. 25. What is a joystick? How is it different from a trackball? Give some typical uses of a joystick. 26. Explain how an electronic pen is used as an input device. 27. Describe the working mechanism of a light pen input device. 28. Describe some of the typical applications for which writing pen with pad is suitable as an input device. 29. What is a Personal Digital Assistant? What type of input device is most suitable for it? 30. What is a touch screen device? Explain the technology used in making touch screen devices. 31. Give some typical applications for which touch screen is most suitable as an input device. 32. What is an information kiosk? Give some typical uses of an information kiosk. 33. Write short notes on: (a) Mouse (b) Trackball (c) Joystick (d) Electronic Pen
(e) Touch screen (f) Information kiosk (g) Personal digital assistants 34. What are data scanning devices? How do they help in improving input data accuracy as compared t< keyboard devices? 35. What is an image scanner? Explain how it inputs a given image document. 36. What is meant by bit map representation of an image? Give the bit map representation of the letter 'A'. Use any suitable grid size for your representation. 37. What is the minimum number of bits required to encode each grid point (pixel) in the bit map representation of an image? How does it help having more number of bits than the minimum requirement? 38. Differentiate between a flat-bed and a hand-held image scanner? 39. What are the limitations of an image scanner when it is used for inputting text documents? How does an OCR device overcomes these limitations? 40. Explain how an OCR device converts the bit map images of characters to equivalent ASCII codes. 41. Explain why an OCR device cannot be successfully used for handwriting recognition. 42. What is an OCR device? List out some of its advantages and limitations for inputting text documents. 43. What is an OMR device? What types of applications is it suitable for? Explain the technique used by it for recognition of marks. 44. Differentiate between an OMR and an OCR device. 45. What are bar codes? How does a bar-code reader read bar-coded data? Give a typical use of a bar-code reader. 46. What is the Universal Product Code (UPC)? Which input device is used for inputting data based on UPC? 47. What is an MICR device? Which industry is the primary user of MICR? 48. Explain how an MICR device helps in faster processing of bank cheques with greater accuracy. 49. What is the main limitation of MICR technology? 50. List out the similarities and differences between an OCR device and an MICR device. 51. What is a digitizer? What are its two main components? What are its main areas of application? 52. Explain how a digitizer can be used to convert a drawing into digital form for storage in computers. 53. What is an electronic card? Give two applications where electronic cards can be effectively used. 54. What is an automatic teller machine (ATM)? Explain how a customer can transact with a bank using an ATM. 55. Explain how electronic cards can be used for controlling access of various categories of people to physically secured areas. 56. What is a smart card? How is it different from a magnetic strip card? 57. Explain how smart cards can be used as electronic money.
58. What is a voice recognition device? Why do currently available voice recognition devices have limited success? 59. Explain how a voice recognition device recognizes and acts upon a spoken input data. 60. Differentiate between speaker-dependent and speaker-independent voice recognition systems. Which one is more difficult to design and why? 61. Give some typical applications of voice recognition systems. 62. What is a vision-input system? Give some typical applications of vision-input systems. 63. Explain how a vision-input device recognizes a given object and takes appropriate action. 64. Differentiate between on-line and off-line data entry devices. 65. What is an off-line data entry device? Why is it used? Name three off-line data entry devices. 66. Explain how data entry operation is carried out using a key-to-tape device. 67. Explain how a key-to-tape or a key-to-floppy device is used to verify the correctness of already entered data. (g) Optical Mark Reader (OMR) (h) Optical Character Recognition (OCR) (i) Automatic Teller Machine (ATM) (j) Voice recognition devices (k) Off-line data entry devices (1) Magnetic-Ink Character Recognition (MICR) (h) Key-to-tape data entry device (i) Key-to-floppy data entry device (j) Key-to-disk data entry device (k) Universal Product Code (UPC) (1) Speaker-dependent voice recognition device (m) Speaker-independent voice recognition device 68. Write short notes on: (a) Image scanner (b) Bar-code reader (c) Digitizer (d) Electronic-card reader (e) Smart Cards (f) Vision-input systems 69. Write short notes on: (a) Bit map representation of images (b) Flat-bed scanner (c) Hand-held scanner (d) OCR software (e) OCR fonts (f) Bar codes (g) Magnetic strip cards 70. Write short notes on: (a) Keyboard devices
(b) Point-and-draw devices (c) Data scanning devices 71. How are graphs arid pictures fed to a computer? 72. What is an output device? Name some of the commonly used output devices. 73. Differentiate between soft-copy and hard-copy outputs. 74. Name two output devices used for producing soft-copy output and two output devices used for producing hard-copy output. 75. What is a computer terminal? Why is it so called? 76. What is a monitor device? Name the two basic types of monitors used today along with their relative advantages and disadvantages. 77. Explain how a CRT monitor displays an output. 78. Why do CRT screens need to be refreshed? 79. Why is the electron beam turned off during horizontal and vertical traces? 80. Why do characters and images displayed by a monitor have to be represented as a pattern of dots? 81. Explain the meaning of the following terms with respect to a CRT monitor: (a) Scan line (e) Interlaced scan (b) Horizontal trace (f) Resolution (c) Vertical trace (g) Refresh rate (d) Progressive scan 82. Differentiate between progressive scan and interlaced scan monitors. Which type of monitor has better quality and why? 83. Differentiate between an alphanumeric display terminal and a graphic display terminal. 84. Explain how a color monitor displays color images. 85. Differentiate among CGA, EGA, VGA and super VGA standards for color monitors; 86. List the following in order of increasing resolution: VGA, CGA, EGA and Super VGA. 87. What is a video RAM? What factors normally decide its capacity? 88. A monitor has a resolution of 800x600. It supports 256 different colors. Calculate the video RAM capacity (in bytes) required by this monitor. 89. What is a flat panel monitor? Where is it commonly used? 90. Explain how LCD monitors display images. 91. List out the relative advantages and limitations of a CRT monitor and a flat-panel monitor. 92. Write short notes on: (a) CRT monitor (b) Video card (c) Video RAM (d) Color monitors (e) Flat-panel monitor (f) Attributes of CRT monitors 93. What is a printer? What are the commonly used types of printers? 94. Differentiate between impact and non-impact printers. Give their relative advantages and disadvantages. Name few printers of both types.
95. What are the units used for measuring printer speed. Also specify which unit is used for what type of printer. 96. What is the unit for measuring printer output quality? According to this feature, what are the common classifications of printers? 97. Can a printer simultaneously support multiple scripts? If no, why? If yes, name few types of printers with this capability. 98. Explain the printing mechanism of dot-matrix printers. 99. List out the key features and limitations of dot-matrix printers. 100. Explain the printing mechanism of inkjet printers. 101. List out the key features and limitations of inkjet printers. 102. List out the relative advantages and limitations of an inkjet printer and a dot-matrix printer. 103. Explain the printing mechanism of drum printers. 104. List out the key features and limitations of drum printers. 105. Explain the printing mechanism of chain/band printers. 106. List out the key features and limitations of chain/band printers. 107. Differentiate between a drum printer and a chain/band printer. List out their relative advantages and limitations. 108. Explain the printing mechanism of laser printers. 109. List out the key features and limitations of laser printers. 110. What is a hydra printer? How is it more useful than a normal printer? 111. Name four types of printers that are used today. Describe the print mechanism of any one of them. 112. Write short notes on: (a) Impact and non-impact printers (b) Printers speed (c) Printers output quality (d) Color printers (e) Dot-matrix printers (f) InkJet printers (g) Drum printers (h) Chain/band printers (i) Laser printers (j) Hydra printers 113. What is a plotter? What type of users need it? 114. Explain the working mechanism of a drum plotter. 115. Explain the working mechanism of a flat-bed plotter. 116. What is a COM device? Explain how information is recorded and retrieved using a COM device. 117. List out the advantages and limitations of a COM system as compared to a printer. 118. For what type of applications is a COM system most suitable? List out some of its typical users. 119. What is a screen image projector? What is it typically used for? 120. What is a voice response system? Give some of its typical applications. 121. What is a voice reproduction system? How does it function? Give some of its typical applications. 122. What is a speech synthesizer? How does it function? Give some of its typical applications.
123. What is a device controller? Explain with a diagram how is it used to connect I/O devices to a computer. 124. Why are I/O devices connected to a computer system through device controllers instead of connecting them directly to the system bus? 125. Differentiate between device controllers with DMA capability and without DMA capability. Which of the two can transfer data faster from an I/O device to the computer's memory and why? 126. What is an I/O port? Differentiate between serial and parallel ports. 127. What is a serial port? Why is-nt so called? Explain how data is transferred from an I/O device to the computer's system bus when the device is connected to a serial port of the system. 128. What is a parallel port? Why is it so called? Explain how data is transferred from an I/O device to the computer's system bus when the device is connected to a parallel port of the system. 129. What is a SCSI interface? Explain how it helps in expanding the capability of a computer system in having more I/O devices connected to it. 130. Define the following terms: (a) SCSI controller (b) SCSI bus (c) SCSI device 131. What is an expansion slot? Explain how a SCSI interface can help in saving expansion slots in a computer system for connecting other devices. 132. What are the advantages of a SCSI controller over a conventional device controller? 133. Differentiate among dumb, smart and intelligent terminals. 134. What is an X terminal? 135. What is a terminal emulator? In what type of scenario is it required? 136. What is a privacy filter? How does it work? 137. What are ergonomically designed devices? Give a few examples of such devices. 138. List out five features of keyboards and monitors that make computer terminals easier to use. 139. Write short notes on: (a) Plotter (b) COM (c) Screen image projector 140. Write short notes on: (a) Device controller (b) I/O ports (c) SCSI interface (d) X terminal (d) Voice response systems (e) Voice reproduction system (f) Speech synthesizer (e) Privacy filter (f) Ergonomically designed devices (g) Dumb, smart and intelligent terminals 141. Write the full forms of the following abbreviations:
(a) GUI (b) CAD (c) OCR (d) OMR (e) UPC (0) MICR (g) VDT (h) CRT (i) COM (j) CGA (k) EGA (1) VGA (m) LCD (n) DMA (o) SCSI (P) DPI (q) CPS (r) LPM (s) NLQ (t) NTQ
Chapter 10 Computer Software The terms hardware and software are frequently mentioned in connection with computers. Hardware is the term given to the machinery itself and to the various individual pieces of equipment. It refers to the physical devices of a computer system. Thus, the input, storage, processing, control, and output devices are hardware. In fact, what we have described so far in the previous chapters is actually the hardware of computer systems. The term "software" will be introduced in this chapter and will be discussed at length in the next few chapters. WHAT IS SOFTWARE? A computer cannot do anything on its own. It must be instructed to do a desired job. Hence it is necessary to specify a sequence of instructions that a computer must perform to solve a problem. Such a sequence of instructions written in a language that can be understood by a computer is called a computer program. It is the program that controls the activity of processing by the computer and the computer performs precisely what the program wants it to do. When a computer is using a particular program, we say it is running or executing that program. The term software refers to the set of computer programs, procedures, and associated documents (flowcharts, manuals, etc.) that describe the programs and how they are to be used. To be precise, software means a collection of programs whose objective is to enhance the capabilities of the hardware. A software package is a group of programs that solve a specific problem or perform a specific type of job. For example, a word-processing package may contain programs for text editing, text formatting, drawing graphics, spelling checking, etc. Thus a multipurpose computer system has several software packages, one each for every type of job it can perform. RELATIONSHIP BETWEEN HARDWARE AND SOFTWARE In order for a computer to produce useful output, the hardware and software must work together. Nothing useful can be done with the computer hardware on its own and software cannot be utilized without supporting hardware. To take an analogy, a cassette player and the cassettes purchased from market are hardware. However, the songs recorded en the cassettes are its software. To listen to a particular song, first of all that song has to be recorded on one of the cassettes which should then be mounted on the cassette player and played. Similarly, to get a particular job done by a computer, the relevant software should be loaded in the hardware before processing starts.
The following important points regarding the relationship between hardware and software are brought out by this analogy: 1. Both hardware and software are necessary for a computer to do useful job. Both are complementary to each other. 2. The same hardware can be loaded with different software to make a computer system perform different types of jobs just as different songs can be played using the same cassette player. 3. Except for upgrades (like increasing the main memory and hard disk capacities, or adding speakers, modems, etc.), hardware is normally a one-time expense whereas software is a continuing expense. Just as we buy new cassettes for the newly released songs or for songs whose cassettes we do not have, similarly one buys new software to be run on the same hardware as and when need arises or funds become available. TYPES OF SOFTWARE A wide variety of computer software is available today. Although the range of software available is vast and varied, most software can be divided into two major categories: 1. System software, and 2. Application software These two types of software along with a few examples of each type are described below. System Software System software is a set of one or more programs designed to control the operation and extend the processing capability of a computer system. In general, a computer's system software performs one or more of the following functions: 1. Supports the development of other application software. 2. Supports the execution of other application software. 3. Monitors the effective use of various hardware resources such as CPU, memory, peripherals, etc. 4. Communicates with and controls the operation of peripheral devices such as printer, disk, tape, etc. Thus system software makes the operation of a computer system more effective and efficient. It he.lps the hardware components work together and provides support for the development and execution of application software (programs). The programs included in a system software package are called system programs and the programmers who prepare system software are referred to as system programmers. System software offers several advantages and conveniences to application programmers and computer users in general. Good system software allows application packages to be run on the computer with less time and effort. Without system software, application packages could not be run on the computer system. Thus system software is an indispensable part of a total computer system. A computer without some kind of system software would be very ineffective and most likely impossible to operate. The production of system software is a complex task that requires extensive knowledge and considerable specialized training in computer science. System programmers are highly trained computer professionals. Because of its technical complexity, system software are normally developed and distributed by the computer manufacturers. The customer who buys or leases a computer system would usually receive, in addition to the hardware, some software needed for the effective operation of his/her computer.
Some of the most commonly known types of system software are operating systems, programming language translators, utility programs, performance monitoring software and communications software. They are briefly described below. Operating Systems Every computer has an operating system software that takes care of the effective and efficient utilization of all the hardware and software components of the computer system. Operating system has been discussed in greater detail in Chapter 14. Programming Language Translators Programming language translators are system software that transform the instructions prepared by programmers using convenient programming languages into a form that can be interpreted and executed by a computer system. Programming language translators have been discussed in greater detail in Chapter 12. Utility Programs Utility programs (also known as utilities) are a set of programs that help users in system maintenance tasks and in performing tasks of routine nature. Some of the tasks commonly performed by utility programs include the following: 1. Formatting hard disks or floppy disks. 2. Reorganizing files on a hard disk to conserve storage space. 3. Taking backup of files stored on hard disk on to a tape or floppy disk. 4. Searching a particular file from a directory of hundreds of files. 5. Checking the amount of available memory. 6. Checking the amount of available storage space on hard disk. 7. Reducing the file size for more efficient transmission over a data communication link. 8. Sorting of the records stored in a file in a particular order based on some key field(s). 9. Scanning the system for computer viruses. Performance Monitoring Software Performance monitoring software (also known as profiler) helps the user in analyzing the performance of the various hardware components as well as the overall performance of a computer system. It provides such information as the percentage of CPU utilization, percentage of memory utilization, and number of disk accesses during any given period of time. Such information can be effectively used by the users in optimizing their programs so that they use the various system resources more optimally and execute faster. It also helps users plan for upgrading the hardware resources of their computer systems such as increasing the CPU power or the memory or hard disk capacity. Communications Software In a network environment (where multiple computers are integrated together by communications network), communications software enables transfer of data and programs from one computer system to another. Typical functions performed by communications software include: 1. Ensuring that connectivity is properly established between the source and destination computer systems that want to exchange information.
2. Encoding the data to be transferred at the source computer in the right format. 3. Transferring the data from the source computer to the destination computer. 4. Decoding the received data at the destination computer. Communications software has been described in greater detail in Chapters 17 and 18. Application Software Application software is a set of one or more programs designed to solve a specific problem or do a specific task. For example, an application software for payroll processing produces pay slips as the major output and an application software for processing examination results produces mark sheets as the major output along with some other statistical reports. Similarly, a program written by a scientist to solve his/her particular research problem is also an application software. The programs included in an application software package are called application programs and the programmers who prepare application software are referred to as application programmers. There are literally millions of application software available for a wide range of applications ranging from simple applications such as word processing, inventory management, preparation «f tax returns, banking, hospital administration, insurance, publishing, to complex scientific and engineering applications such as weather forecasting, space shuttle launching, oil and natural gas exploration, design of complex structures like aircrafts, ships, bridges, sky-rise buildings, etc. With so many applications available, it is not possible to categorize them all and to cover them here. Hence just to have a feel of what application software does, some of the most commonly known application software are briefly described below. Word-Processing Software Word-Processing is a term that describes the use of computers to create, edit, view, format, store, retrieve and print documents (written material such as letters, reports, books, etc.). A word-processing software is an application software that enables the user to do all these on a computer system. The need to create documents is so common in every walk of life, whether it is at work, at school, or at home, that word-processing software is the most commonly used application software. Word-processing software has been discussed in greater detail in Chapter 15. Spreadsheet Software A spreadsheet software is a numeric data analysis tool that allows you to create a kind of computerized ledger. A manual ledger is a book having rows and columns that accountants use for keeping a record of financial transactions and for preparing financial statements. Accountants use the manual ledger with pencil, erasure and hand calculator to prepare financial statements. This is a tedious task and often takes a long time due to several iterations of formula calculations to come out with an acceptable and satisfactory financial statement. A spreadsheet software offers considerable ease of performing such tasks by automating all arithmetic calculations and making it much easier to change certain numeric values and immediately seeing the effect of these changes across the worksheet (ledger). With spreadsheet software in place, we are no longer confined to using pencils, erasers, and hand calculators for dealing with any task that requires
numeric data analysis of the nature mentioned above. Spreadsheet software has been discussed in greater detail in Chapter 15. Database Software This application software is designed to take advantage of the tremendous capability of computer systems in storing, organizing and selectively retrieving large volumes of data efficiently. A database is a collection of related data stored and treated as a unit for information retrieval purposes. It is important here to note the difference between data and information. Data is stored in a database whereas information is retrieved from a database. That is, data is input to any data processing system and information is its output. In other words, data is the raw material for a data processing system and information is the processed data. A database concept can be best illustrated with an example. For example, a school's student database might contain the name, sex, date of birth, current address, permanent address, parent's occupation, marks scored in various subjects in each class already passed, and current occupation for all the students who are either currently enrolled or have already graduated from the school. Similarly, an individual's address database might contain the names, addresses, and telephone numbers of all his/her friends and business contacts. The data of one student in the student database or one person in the address database is known as a record. A database software is a set of one or more programs that enables users to create a database, maintain it (add, delete and update its records), organize its data in desired fashion (for example, sort its records alphabetically name-wise), and to selectively retrieve useful information from it (for example, get the telephone number of the person named Kashyap Rana from the address database, or get the names of all currently enrolled students from the student database whose birthdays fall today). Database software is widely used today by both large and small organizations to maintain employee data, customer data, assets data, accounting data, etc. Even individuals find the database software very useful for maintaining data such as list of addresses and telephone numbers of their friends and relatives, list of household goods and other valuables, list of various types of savings in different banks, etc. Database software has been discussed in greater detail in Chapter 16. Graphics Software A graphics software enables us to use a computer system for creating, editing, viewing, storing, retrieving and printing designs, drawings, pictures, graphs and anything else that can be drawn in the traditional manner. Personal Assistance Software A personal assistance software allows us to use personal computers for storing and retrieving our personal information, and planning and managing our schedules, contacts, financial and inventory of important items. Education Software Education software allow computer systems to be used as a teaching and learning tool. A few examples of such applications are:
1. Applications that teach young children to do mathematics, how to recognize the alphabet, or how to read whole words and sentences. The use of a computer allows the learning process to be made interesting by incorporating audio-visual effects in the application. For example, use of encouraging spoken words (like excellent, marvelous, etc.) when the correct answer is given, and display of marks obtained leads to interesting learning process. 2. Applications that allow students to prepare for tests and entrance examinations. These applications simulate a test or an examination by asking similar questions to the student and guide the student when he/she answers wrongly to a question. The beauty of these applications is that they are designed to prepare each student at his/her own knowledge level and pace of learning. That is, the questions become tougher at the pace with which a student provides correct answers. So the questions will soon get tougher for a brilliant student and will slowly get tougher for an average or below average student. Another interesting and useful feature of these applications is that they can immediately analyze the type of questions in which a student is having difficulty and can put forth more questions of that type to the student so that he/she gets more practice in his/her weak areas. 3. Applications that allow an individual to take on-line examinations in which an examinee has to use a computer during the examination. The computer presents questions one at a time on its screen and the examinee has to click the right answer from multiple choices presented to him/her. In some online examinations each question has a difficulty level attached to it. It means that the examination automatically adapts itself according to how the student answers the previous question. If he/she answers a question correctly, he/she gets a more difficult question and vice-versa. Due to the availability of such applications, today more and more examinations are being conducted on-line. A few examples are GRE (Graduate Record Examination which is an aptitude test conducted by the United States of America for those who want to seek admission in American Universities), tests for obtaining certificates as a qualified professional in a certain area (like the tests for becoming Microsoft Certified Professionals in various areas of skill), etc. The beauty of these applications is that the result of the examination (pass/fail and percentage of marks obtained) is declared to the examinee immediately after the examination is over (before the examinee leaves the computer that he/she was using for the examination). This is because the computer dynamically keeps updating the score as the examinee progresses through the examination questions. 4. Applications that allow individuals to learn foreign languages. Use of such applications has been found to be very useful as compared to the use of books because these applications incorporate both text and sound to allow the learner to see words and phrases on the screen as they are pronounced by the computer. The learner als9 has the flexibility to request the computer to pronounce any word displayed on the screen by simply selecting that word (say by clicking over it). Often these applications also have additional features such as if the learner does not understand the meaning of a word or phrase, he/she may point to it to quickly display a translation for it in his/her native language. 5. Applications that allow hard-copy books and other reading material to be made available in soft-copy form on computers. In these applications, the power of computer is effectively utilized for quickly searching for specific facts of interest to a particular
reader. Due to this feature, several reference books such as encyclopedias and dictionaries are now being published and made available on CD-ROMs. Many journals, magazines and product manuals are also being published in electronic form so that users can access and print only those articles/portions in which they are interested. A further extension of this idea is digital library, which is a repository of large information in digital form allowing multiple users to search and access information of their interest from it. Entertainment Software Entertainment software allow computer systems to be used as an entertainment tool. A few examples of such applications are: 1. Video Games. Today thousands of video game applications are available on computer systems. Special audio and visual effects are used in these applications to make the games thrilling and exciting for the users. 2. Interactive Television. These applications allow users to either use their computer systems as an interactive television or to convert their conventional televisions into an interactive television. An interactive TV allows users to select TV programs of their interest from a massive video archive, view at the time they wish, and interact with the programs via VCR-like functions, such as fast-forward and rewind. In essence, it implements a virtual video-rental shop from which customers can borrow and view the videos of their choice at their own convenient time. The rental shop being virtual provides 24 hours service. 3. Game Partner. These applications allow computers to become a game partner of an individual for playing conventional games that require two players. A good example of such an application is the use of computers for playing chess. LOGICAL SYSTEM ARCHITECTURE The logical architecture of a computer system is shown in Figure 10.1. The architecture basically depicts the relationship among the hardware, system software, application software and users of a computer system. As shown in the figure, at the center of any computer system is the hardware, which comprises of the physica devices/components of the computer system. Surrounding the hardware is the system software layer that constitutes the operating and programming environment of the computer system. That is, the software at this layer is designed to hide the hardware details of the system from application programmers and to coordinate the operations of the various hardware devices for optimizing the performance of all the devices. Surrounding the system software is the application software layer that consists of a wide range of software that are designed to do; specific task or solve a specific problem. The final layer is the layer of users who normally interact with the system via the user interface provided by the application software. Different application software usually provides different user interfaces. Hence how a particular user interacts with the computer system depends on whicl application he/she is using. . -
HARDWARE (Physical devices/components of the computer system) SYSTEM SOFTWARE (Software that constitute the operating and programming environment of the computer system) APPLICATION SOFTWARE (Software that do a specific task or solve a specific problem) USERS (Normally interact with the system via the user interface provided by the application software) Figure 10.1. Relationship among the hardware, system software, application software, and users of a computer system. ACQUIRING SOFTWARE Computer software has evolved from an item of minor economic concern with first generation digital computer to one of major concern in today's computing systems. In the early days of computers, more than 80% of the tots computer cost was devoted to hardware, but today more than 80% of the cost is for software. At one time application and system software were included in the purchase price of the computer. Today, however, software is usually not included in the purchase price of the computer. For most computer manufacturers the purchase price of a computer includes only the hardware along with a minimum of system software. The customs normally has to pay extra charges for additional system software and application software that he/she may wish 1 purchase. A desired software may be obtained today in one or more of the ways discussed below. The relative advantage and limitations of each way of obtaining software are also discussed below. A user must make a careful analysis of his/her requirements and the relative advantages and limitations mentioned below before deciding on a particular option for obtaining desired software. Buying Pre-written Software Thousands of pre-written software packages are available today. If you can find a software package that meets your requirements, purchasing it is probably the best option. The following steps are typically followed in selecting and buying a pre-written software package by a user: 1. The user must first prepare a list of all available software packages that are meant for performing the task(s) for which the user is looking for a software. Some of the commonly used sources of information for preparing the lisj are advertisements and reviews of software products published in computer magazines, and demonstrations in retain stores that sell software packages. 2. The second step is to select only those software packages from the list prepared above that will meet the requirements of the user. For example, not all software packages run on all computers. So the user must select only those software packages that are compatible with the available/planned hardware. Other factors that are commonly used
to further narrow down the selection list are the input data format accepted, the format and contents of the output produced, the range of I/O devices supported, the file formats supported for storage, the average response time, and the quality of the documents supported by the software package. This step will normally reduce the list of selected software packages to only a few candidates. 3. The third step is to choose the best one from the list of selected candidates. The factors commonly used for this purpose are actual demonstration of the usage of the package on user's system with his/her data, the duration and level of warranty support provided by the software vendor, and the cost of the package as compared to other selected candidates. 4. The final step is to find out the source from where the finally chosen software can be purchased at the cheapest price. Possible sources for buying a software package are directly from the producer, from a retail store, or from a mail-order distributor. Different vendors normally offer different discount rates on the list price and selecting the best vendor in terms of price and after-sale support is very important. The following are the advantages and limitations of buying a pre-written software package: 1. A pre-written software package usually costs less because its development and maintenance costs are shared by many customers. 2. With a pre-written software package, a user can start the planned activity almost immediately. The user need not wait for the software to be developed and tested before the planned activity can be started. This may be very important if the development and testing efforts involve several months. 3. Pre-written software packages are usually designed to be general purpose so that they can meet the requirements of as many potential users as possible. Due to this feature, many times the operating efficiency and the capability to meet the specific needs of a user more effectively is not as good for pre-written software packages as for in-house developed software packages. Ordering Customized Software Often there are situations where none of the available pre-written software packages can meet the specific requirements of a user (an organization or an individual) either partially or fully. In such a situation, it becomes necessary to create a customized software package to satisfy the specific requirements. If the user has an in-house software development team, the software package can be created in-house. However, if such a team does not exist in-house, the user must get it created by another organization by placing an order for it. The following steps are typically followed for this: 1. The user has to first of all list out all its requirements very carefully. 2. The user then floats a tender for inviting quotations for the creation of the requisite software. Sometimes the user may directly contact few software houses instead of floating a tender for quotations. 3. After the quotations are received, the user shortlists a few of them for further interaction based on the cost quoted by them, their reputation in the market, their submitted proposal, etc.
4. The user then personally interacts with the representative(s) of each of the shortlisted vendors. Based on this interaction, the user makes a final choice of the vendor whom it wants to offer the contract of creation of the requisite software. 5. The selected vendor then creates the software package and delivers it to the user. Often the vendor has to very closely interact with the user during the software development process. Note that the vendor need not develop everything from scratch to meet the user's requirements. If one or more pre-written software packages partially meet the user's requirements, the vendor may as well use them along with its own developed software in such a manner that the overall software package so created meets the user's requirements fully. This will greatly help in reducing the overall development cost and time needed for the creation of the software package. Often the user has to order for both the hardware and the software. In this case, the user may choose to place the order for both to a single vendor. The vendor develops the software on the chosen hardware and delivers the software along with the hardware to the user. This is normally referred to as an end-to-end solution or a turnkey solution. The following are the advantages and limitations of ordering a customized software package rather than developing it in-house: 1. In case of ordering, the user need not maintain its own software development team. Maintaining and managing such a team is an expensive affair and may not be justified for an organization that does not need to develop software on a regular basis. 2. It is easier to carry out changes in the software if it is developed by an in-house team. For ordered customized software, the user needs to always depend on the vendor for carrying out the changes and the vendor may separately charge for every request for change. Developing Customized Software If none of the available pre-written software packages can meet the specific requirements of an organization and if the organization has an in-house software development team, the organization may very well choose to get; customized software package developed inhouse to satisfy the specific requirements. The following steps ar typically followed for the in-house development of a software package: 1. A project team is first constituted to carry out the development activity. 2. The team studies the requirements carefully and plans the functional modules for the software. 3. It then analyzes which of the functional modules need to be developed and which of the functional modules' requirements can be met with an existing pre-written software. 4. For the functional modules that need to be developed, the team next plans their programs and does the coding, testing, debugging and*documentation for the planned programs. 5. All the modules are then tested in an integrated manner. 6. The software is then implemented, used and maintained. The following are the advantages and limitations of developing a customized software package in-house rather than getting it developed by an outside party: 1. It is easier to carry out changes in the software if it is developed in-house.
2. Developing software in-house means a major commitment of time, money and resources because an in-house software development team needs to be maintained and managed. Downloading Public-domain Software A public-domain software is a software that is available for free or for a very nominal charge from the bulletin boards or user-group libraries on the Internet. Creators of these software obtain distribution for their products by supplying free copies to bulletin boards and user-group libraries. Their basic idea behind this is to popularize their software to as many users as possible. Users are encouraged to copy these software and try them out. The software can be freely shared with other users. Hence public-domain software are also referred to as shareware or user-supported software. Often a user may find a public-domain software suitable for his/her requirements. In this case, he/she can obtain it for free by downloading it from the Internet. The following are the advantages and limitations of downloading and using publicdomain software packages: 1. They are available for free. 2. They can be downloaded and used immediately. The user need not wait for the software to be developed and tested before the planned activity can be started. 3. They may not be properly tested before release and their support is normally poor as compared to a commercial software. Hence they may fail during operation and bug fixing may not take place soon. SOFTWARE DEVELOPMENT STEPS No matter whether a software is a pre-written software, or a customized software developed by a vendor or in-house, or a public-domain software, it has to be developed by someone in the first place. Developing a software and putting it to use is a complex process and involves the following steps: 1. Analyzing the problem at hand and planning the program(s) to solve the problem. 2. Coding the*program(s). 3. Testing, debugging and documenting the program(s). 4. Implementing the program(s). 5. Evaluating and maintaining the program(s). These steps have been covered in detail in the next few chapters. Chapter 11 deals with Step 1, Chapter 12 deal: with Step 2, and Chapter 13 deals with Steps 3, 4 and 5. SOFTWARE ENGINEERING What is Software Engineering? Towards the beginning of this chapter we saw that "software" is defined as the set of computer programs procedures, and associated documents that describe the programs and how they are to be used. On the other hand, "engineering" is defined as the systematic application of scientific knowledge in creating and building cost-effective solutions to practical problems. Hence "software engineering" is that branch of engineering that deal
with the systematic application of the principles of computer science and mathematics in creating and building cost-effective software solutions.
Need for Software Engineering We saw that software has evolved from an item of minor economic concern with first generation digit; computers to one of major concern in today's computing systems. With the increased use of computers in ever walk of life, the volume of software required has steadily increased, and the costs of obtaining this software ha\ grown even more rapidly. With the larger-scale software products, the number of development people involved i a software project has proportionately increased. It was found that managing the development of large software products and maintaining them was becoming an increasingly difficult task. The software projects were frequently faced with the problems of missed schedules, cost overruns, and low quality software products, which were difficult to maintain. Moreover, the use of progressively larger software products in sensitive application where errors in software products could have severe and even lifethreatening consequences required t\ correctness and reliability of software products to be of very high degree. These quality and productivity demamds for software products led to the introduction of systematic practices (later on known as software engineering practices) in the software industry. Goals of Software Engineering The goals of software engineering are to produce software products that have the following properties: 1. Their correctness should be of very high degree. Correctness refers to the degree to which the software product performs its intended, function properly, consistently, and predictably. 2. Their usability should be of very high degree. Usability refers to the ease with which the software product and its associated documentation are usable by the intended audience. 3. They should be cost-effective. Cost-effectiveness means that the total development and operational costs of a software product should be as low as possible without compromising on the quality of the product in any way. Note that overall costeffectiveness also requires that a software product is capable of being readily modified and enhanced. It has been found that software products that meet these goals do not generally cause unusual stress during their various phases of development and are easily adaptable to unanticipated needs. Principles of Software Engineering The principles of software engineering are the generally accepted guidelines that should be followed to produce a software product for meeting the goals of software engineering. These guidelines are described below. Precise Requirements Definition
The requirements of a software product must be precisely defined to ensure that the resulting product meets the users' true needs. Ill-defined and fluid nature of software requirements generally lead to problems in the later phases of the software life cycle. Modular Structure Software systems should be structured in a modular fashion so that they are easy to program, implement and maintain. A modular design helps in distribution of different modules to multiple programmers allowing the product to be developed in a shorter time. It also helps in easier testing, debugging and maintenance of software products because individual modules can be independently tested, debugged and maintained. Feature enhancements can be done by adding new modules. Modular design also enables software reusability because if a module already exists for a desired functionality, it can be used as it is or with minor customization. Third party library routines are an example of reusable software modules. Abstraction As far as practicable, software systems should use abstraction and information hiding. For example, in modular design, the implementation details of each module should be hidden inside the module and only the module interface should be made visible outside to allow other modules to interact with the module. The concept of object-oriented programming has been introduced to take care of this aspect of software engineering. Abstraction helps in easy reusability of existing modules because for reusing a module only its interface needs to be understood and not its internal details. Uniformity Software systems should maintain uniformity in design, coding, documentation, etc. Uniformity ensure; consistency, which makes the software product easier to develop and maintain as well as easier to learn and use For example, if all the graphical user interfaces of a software product follow a uniform style (providing the same look-and-feel throughout), it becomes much easier for a user to learn and use the software. Similarly, if uniform coding and documentation styles are followed by all the programmers working on the various modules of software project, it becomes much easier for someone else to understand and maintain the modules. Several software engineering standards, such as design standards, coding standards, documentation standards, etc. have been introduced to take care of this aspect of software engineering. CASE Tools Proper software tools should be used in the various phases of a software product's life cycle. As in other fields of engineering, software tools play an important role in software engineering. They help in increasing professional capability and eliminate or reduce sources of error in software products. Over the years, a large number o software tools have been made available to aid the software developers. In fact, the popularity of these tools resulted in the creation of a new area called CASE (Computer Aided Software Engineering). CASE tools provide a wide range of features for creation of better and
more reliable software products within less time and with less effort. A few of the most common features are: 1. Design specification tools allow programmers to visually design screens, menus, database tables reports, dialog boxes, and other key components of a program. With the individual component defined, the programmer can link the components together and then test the operation of the program without writing even a single line of program code. 2. Code generation tools generate source code (programs) from the design specification of a system. That is, once satisfied with the structure, design, and components of the system, the programmer can invoke a code generator, which creates the source code for the program automatically. Note that this automatic creation of programs by the system leads to error-free program codes. 3. Testing and debugging tools help programmers in testing and debugging of their programs a already described before in this chapter. 4. Source-code analysis tools help in optimizing a program by pointing out unreachable lines of cod and functions that are never called. They also assist in carrying out program changes consistently and uniformally by indicating what other parts of the program would be affected by introducing a change in the program. 5. Documentation tools assist in automatic generation of technical documents for software systems For example, automatic flowchart generation tools can create flowcharts for already written programs Similarly, as a programmer uses a design specification tool to design the screens, menus, reports dialog boxes, and other key components of a program, the documentation tool collects this information for automatically producing the system's documentation. With good CASE tools, much of the tedious detail work of designing and writing a program and writing documents for it can be drastically reduced. Hence CASE tools can save several hours of time to get a new program up and running, allowing the programmer to concentrate on the important functional parts of a program, rather than spending time writing the code to display screens, dialog boxes, and menus. CASE tools, however, are not the answer for every situation, and even in systems that make extensive use of CASE tools during their production, some hand coding and manual documentation is usually necessary. FIRMWARE Computer software in conventional systems is supplied on storage media like CDs, floppies, tapes, disks, etc. However, with the advancement in technology and the reduction in hardware cost, today software is also being made available by many computer manufacturers on read-only memory (ROM) chips. These ROM chips can be easily plugged into the computer system and they form a part of the hardware. Such programs (software) made available on hardware are known as firmware. Firmware often refers to a sequence of instructions (software) that is substituted for hardware. For example, when cost is more important than performance speed, the computer system architect might decide not to use special electronic circuits (hardware) to multiply two numbers, but instead write instructions (software) to cause the machine to accomplish the same function by repeated use of circuits already designed to perform addition. This software will be stored in a ROM chip of the computer system and will be executed (used) whenever the computer has to multiply two numbers. Hence, this software will be
known as firmware. To be precise, firmware is software substituted for hardware and stored in read-only memory. Initially, only systems software was supplied in the form of firmware. But today, even application programs are being supplied in firmware form. Dedicated applications are also programmed in this fashion and available in firmware. Because of the rapid improvements in memory technology, firmware is frequently a cost-effective alternative to wired electronic circuits, and its use in computer design has been gradually increasing. In fact, the increased use of firmware has today made it possible to produce smart machines of all types. These machines have microprocessor chips with embedded software. Points to Remember 1. The term hardware refers to the physical devices of a computer system. Thus, the input, storage, processing, control, and output devices are hardware. 2. The term software refers to a collection of programs. A program is a sequence of instructions written in a language that can be understood by a computer. It is the program that controls the activity of processing by the computer and the computer performs precisely what the program wants it to do. 3. A software package is a group of programs that solve a specific problem or perform a specific type of job. 4. The hardware and software of a computer must work together for the computer to produce useful output. Nothing useful can be done with the computer hardware on its own and software cannot be utilized without supporting hardware. 5. Most software can be divided into two major categories - system software and application software. System software is a set of one or more programs designed to control the operation and extend the processing capability of a computer system. Application software is a set of one or more programs designed to solve a specific problem or do a specific task. 6. A few examples of system software are operating systems, programming language translators, utility programs, performance monitoring software, and communications software. 7. A few examples of application software are word-processing software, spreadsheet software, database software, graphics software, personal assistance software, education software, and entertainment software. 8. Figure 10.1 shows the relationship among the hardware, system software, application software, and users of a computer system. 9. A user can acquire a desired software in or more of the following ways: • By buying pre-written software • By ordering customized software • By developing customized software • By downloading public-domain software Each of these ways of acquiring software has its own advantages and limitations. 10. Developing a software and putting it to use is a complex process and involves the following steps: • Analyzing the problem at hand and planning the program(s) to solve the problem. • Coding the program(s).
• Testing, debugging and documenting the program(s). • Implementing the program(s). • Evaluating and maintaining the program(s). 11. Software engineering is that branch of engineering that deals with the systematic application of the principles of computer science and mathematics in creating and building cost-effective software solutions. 12. Firmware is software substituted for hardware and stored in read-only memory. Questions 1. Define the terms hardware and software. 2. What is a computer program? 3. What is meant by running/executing a computer program? 4. What is a software package? 5. Hardware and software of a computer system are like two sides of a coin. Discuss. 6. Give an analogy to bring out the relationship between hardware and software of a computer system. 7. Hardware is normally a one-time expense whereas software is a continuing expense. Elaborate. 8. How many types of software are there? Give three examples of each. 9. Define and distinguish between application software and system software. 10. Define the following terms: (a) System software (b) System program (c) System programmer (d) Application software (e) Application program (f) Application programmer 11. List out some of the key functions performed by the system software of a computer system. 12. Explain the importance of system software for a computer system. 13. What is a utility program? List out some of the tasks commonly performed by utility programs. 14. What is a profiler? How is it useful for computer users? 15. What is a communications software? What are some of its typical functions? 16. What is a database? Give an example of a database application. 17. What is a database software? List out some of its typical uses. 18. Differentiate between data and information. 19. What is an education software? Give a few examples of such software. 20. What is an entertainment software? Give a few examples of such software. 21. Explain the relationship among the hardware, system software, application software, and users of a computer system. 22. How does a normal user interact with the hardware of a computer? Describe the various in-between layers. 23. What are the different ways of acquiring software? List out their relative advantages and limitations. 24. List out the steps typically followed by a user in selecting and buying a pre-written software package.
25. List out the steps typically followed by a user in placing an order on another organization for the creation of a customized software. 26. List out the relative advantages and disadvantages of creating a customized software in-house by an organization versus getting it created by another organization. 27. What is a turnkey solution? 28. List out the steps typically followed in the in-house development of a software package. 29. What is a shareware? What are the advantages and limitations of using a shareware? 30. List out the steps typically followed in developing a software and putting it to use. 31. What is software engineering? Why is it needed? 32. List out the main goals of software engineering. 33. List out the main principles of software engineering. Explain how each of these principles helps in achieving one or more goals of software engineering. 34. What is meant by modular structure of a software system? How does it help? 35. What is meant by abstraction in software systems? How does it help? 36. Why should software systems maintain uniformity in design, coding, documentation, etc.? How can uniformity be ensured? 37. What are CASE tools? Hew are they useful? 38. List out some commonly used CASE tools? Explain how each one of them is useful. 39. What is a source-code generator? How is it useful? 40. What is a source-code analyzer? How is it useful? 41. CASE tools are not the answer for every situation. Discuss. 42. Write short notes on: (a) Hardware (b) Software (c) Firmware 43. Write short notes on: (a) Types of software (b) Acquiring software (c) Software engineering (d) Software development steps (e) Relationship between hardware and software 44. Write short notes on: (a) System software (b) Application software (c) Principles of software engineering (d) Relationship among the hardware, system software, application software, and users of a computer 45. Write short notes on: (a) Utility programs (b) End-to-end solution (c) Shareware (d) CASE tools (e) Education software (f) Entertainment software (g) Communications software
(h) Performance monitoring software 46. What is firmware and what is its importance to the computer system architect? 47. Why is firmware gaining popularity? 48. Differentiate among hardware, software, and firmware.
Chapter 11 Planning the Computer Program In Chapter 10, computer software has been described as a set of computer programs. We also saw in Chapter 10 that before writing a program for a software, one must first plan the program. That is, to produce a correct and effective computer program, one must first plan the logic (the various steps) of the program. If one attempts to plan the logic and write the program at the same time, he/she will likely become so involved with the required instruction formats that program logic will suffer. Hence, before we learn how to write a program (next chapter), we will first learn how to plan the logic of a computer program in this chapter. PURPOSE OF PROGRAM PLANNING Suppose you are asked by your teacher to solve an arithmetic problem and you are not familiar with the steps involved in solving that problem. In such a situation, you will not be able to solve the problem. The same principle applies to writing computer programs also. A programmer cannot write the instructions to be followed by a computer unless the programmer knows how to solve the problem manually. Suppose you know the steps to be followed for solving the given problem but while solving the problem, you forget to apply some of the steps or you apply the calculation steps in the wrong sequence. Obviously, you will get a wrong answer. Similarly, while writing a computer program, if the programmer leaves out some of the instructions for the computer or writes the instructions in the wrong sequence, then the computer will calculate a wrong answer. Thus, to produce an effective computer program, it is necessary that the programmer write each and every instruction in the proper sequence. However, the instruction sequence (logic) of a computer program can be very complex. Hence, in order to ensure that the program instructions are appropriate for the problem and are in the correct sequence, programs must be planned before they are written. ALGORITHM What is an Algorithm? We saw above that planning a program basically involves defining its logic (the correct sequence of instructions needed to solve the problem at hand). The term algorithm is often used to refer to the logic of a program. It is a step-by-step description of how to arrive at the solution of the given problem. It may be formally defined as a sequence of instructions designed in such a way that if the instructions z executed in the specified sequence, the desired results will be obtained. The instructions, however, should precise and unambiguous and the result should be obtained after a finite number of executional steps. The latter condition actually states that an algorithm must terminate and should not repeat one or more instructions infinitely. Hence, in order to qualify as an algorithm, a sequence of instructions must possess the following characteristics:
1. Each and every instruction should be precise and unambiguous. 2. Each instruction should be such that it can be performed in a finite time. 3. One or more instructions should not be repeated infinitely. This ensures that the algorithm w ultimately terminate. 4. After performing the instructions, that is after the algorithm terminates, the desired results must obtained. Sample Algorithms To gain insight into algorithms, let us consider some simple examples. Example 11.1. There are 50 students in a class who appeared in their final examination. Their mark sheets have bi given to you. The division column of the mark sheet contains the division (FIRST, SECO> THIRD or FAIL) obtained by the student. Write an algorithm to calculate and print the total num of students who passed in FIRST division. Algorithm: Step 1: Initialize Total_First_Division and Total_Marksheets_Checked to zero. Step 2: Take the mark sheet of the next student. Step 3: Check the division column of the mark sheet to see if it is FIRST, if no, go to Step 5. Step 4: Add 1 to Total_First_Division. Step5: Add 1 to Total_Marksheets_checked. Step 6: Is Total_Marksheets_Checked = 50, if no, go to Step 2 Step 7: Print Toral_First_Division. Step 8: Stop. Example 11.2. There are 100 employees in an organization. The organization wants to distribute annual bonus to the employees based on their performance. The performance of the employees is recorded in their annual appraisal forms. ""Every employee's appraisal form contains his/her basic salary and the grade for his/her performance during the year. The grade is of three categories - 'A' for outstanding performance, 'B' for good performance and 'C for average performance. It has been decided that the bonus of an employee will be 100% of the basic salary for outstanding performance, 70% of the basic salary for good performance, 40% of the basic salary for average performance, and zero for all other cases. Write an algorithm to calculate and print the total bonus amount to be distributed by the organization. Algorithm: Step 1: Initialize Total_Bonus and Total_Employees_Checked to zero. Step 2: Initialize Bonus and Basic_Salary to zero. Step 3: Take the appraisal form of the next employee. Step 4: Read the employee's Basic_Salary and Grade. Step 5: If Grade = A, then Bonus = Basic_Salary. Go to Step 8. Step 6: If Grade = B, then Bonus = Basic_Salary x 0.7. Go to Step 8. Step 7: If Grade = C, then Bonus = Basic_Salary x 0.4. Step 8: Add Bonus to Total_Bonus. Step 9: Add 1 to Total_Employees_Checked. Step 10: If Total_Employees_Checked < 100, then go to Step 2.
Step 11: Print Total_Bonus. Step 12: Stop. Programmers design algorithms like these to solve complex programming problems. It must be clear to the readers from these examples that even for very simple problems, the development of algorithms is not so simple as it might initially appear and requires some thinking. It may also be noted from the given examples that in order to solve a given problem, each and every instruction must be strictly carried out in a particular sequence. It is this fact, which a beginner to problem solving by computers finds difficult to appreciate. Quality of Algorithms Given a solvable problem, there are often many methods to solve it. All of these methods may not be equally good. Similarly, for a given problem, there may be many algorithms, not all of equal quality. The primary factors that are often used to judge the quality of an algorithm are: 1. Time requirement. This is the time required to execute the corresponding program on a given computer system. The lesser is the time requirement, the better is the algorithm. 2. Memory requirement. This is the memory space required to execute the corresponding program on a given computer system. The lesser is the memory requirement, the better is the algorithm. 3. Accuracy of solution. Although multiple algorithms may provide correct solutions to a given problem, some of these may provide more accurate results than others. For example, out of two algorithms for calculating the square root of a given number, one might have been designed to produce results up to three decimal places, while the other one might have been designed to produce results up to six decimal places. Obviously, the latter algorithm is better from the point of view of accuracy of solution. 4. Generality. While there are occasions when an algorithm is needed to solve a single isolated problem, more often algorithms are designed to handle a range of input data. A generalized algorithm, which can handle a range of input data, is better than one, which has been designed to solve a problem for a single input data. For instance, the algorithm given above for Example 11.1 is not a generalized one because in Step 6 it checks if Total_Marksheets_Checked has become equal to 50. Hence it will work only if there are exactly 50 students data as input. The algorithm can be made general by accepting the number of students (iV) as input and then comparing the value of Total_Marksheets_Checked with JV instead of 50 in Step 6. Representation of Algorithms There are various ways in which an algorithm can be represented. Programmers normally use one or more of the following ways to represent their algorithms: 1. As programs 2. As flowcharts 3. As pseudocodes 4. As decision tables
When an algorithm is represented in the form of a programming language, it becomes a program. Thus any program is an algorithm although the reverse is not true. Besides represented as programs, algorithms are often represented as flowcharts, pseudocodes and decision tables. These are the commonly used tools used by programmers for program planning which can often be of immense help in developing an effective and correct algorithm for a program before it is coded into a programming language. These program planning tools are described below. Programmers generally select one of these tools FLOWCHARTS What is a Flowchart? A flowchart is a pictorial representation of an algorithm. It is the layout, in a visual, twodimensional format, of the plan to be followed when the corresponding algorithm is converted into a program by writing it in a programming language. It acts like a roadmap for a programmer and guides him/her on how to go from the starting point to the final point while converting the algorithm into a computer program. A flowchart is often used by programmers "as a program planning tool for organizing a sequence of steps necessary to solve a problem by a computer. It uses boxes of different shapes to denote different types of instructions. The actual instructions are written within these boxes using clear and concise statements. These boxes are connected by solid lines having arrow marks to indicate the flow of operation, that is, the exact sequence in which the instructions are to be executed. The process of drawing a flowchart for an algorithm is often referred to as flowcharting. Why Use Flowcharts? Normally, an algorithm is first represented in the form of a flowchart and the flowchart is then expressed in some programming language to prepare a computer program. The main advantage of this two step approach in program writing is that while drawing a flowchart, a programmer is not concerned with the details of the elements of programming language. Hence, he/she can fully concentrate on the logic of the procedure. Moreover, since a flowchart shows the flow of operations in pictorial form, any error in the logic of the procedure can be detected more easily than in the case of a program. Once the flowchart is ready, the programmer can forget about the logic and can concentrate only on coding the operations in each box of the flowchart in terms of the statements of the programming language. This will normally ensure an error-free program. Experienced programmers sometimes write programs without drawing the flowchart. However, for a beginner it is recommended that a flowchart be drawn first in order to reduce the number of errors and omissions in the program. Moreover, it is a good practice to have a flowchart along with a computer program because a flowchart often serves as a document for the computer program and is very useful during the testing of the program as well as while incorporating further modifications in the program.
Flowchart Symbols Need for Flowchart Symbols We have seen that a flowchart uses boxes of different shapes to denote different types of instructions. The communication of program logic through flowcharts is made easier through the use of symbols that have standardized meanings. This is because as long as everyone uses the same basic shapes, others can readily interpret the logic. For example, a diamond always means a decision. Hence when a programmer looks at a flowchart, he/she can easily identify the decision points because all are indicated by a diamondshaped box. Basic Flowchart Symbols Only a few symbols are needed to indicate the necessary operations in a flowchart. These basic flowchart symbols have been standardized by the American National Standards Institute (ANSI). These symbols are shown in Figure 11.1 and their functions are discussed below. Terminal Input/Output Processing Decision Flow lines Connectors 1. Terminal. The terminal symbol, as the name implies, is used to indicate the beginning (Start), ending (Stop), and pauses (Halt) in the program logic flow. It is the first symbol and the last symbol in the program logic. In addition, if the program logic calls for a pause in the program, that also is indicated with a terminal symbol. A pause is normally used in the program logic under some error conditions or if forms had to be changed in the computer's line printer during the processing of that program. 2. Input/Output. The input/output symbol is used to denote any function of an input/output device in the program. If there is a program instruction to input data from a disk, tape, terminal, or any other type of input device, that step will be indicated in the flowchart with an input/output symbol. Similarly, all output instructions, whether it is output on a printer, magnetic tape, magnetic disk, terminal screen, or any output device, are indicated in the flowchart with an input/output symbol. 3. Processing. A processing symbol is used in a flowchart to represent arithmetic and data movement instructions. Thus, all arithmetic processes of adding, subtracting, multiplying and dividing are shown by a processing symbol. The logical process of moving data from one location of the main memory to another is also denoted by this symbol. When more than one arithmetic and data movement instructions are to be executed consecutively, they are normally placed in the same processing box and they are assumed to be executed in the order of their appearance. 4. Decision. The decision symbol is used in a flowchart to indicate a point at which a decision has to be made and a branch to one of two or more alternative points is possible. Figure 11.2 shows three different ways in which a decision symbol can be used. It may be
noted from these examples that the criterion for making the decision should be indicated clearly within the decision box. Moreover, the condition upon which each of the possible exit paths will be executed should be identified and all the possible paths should be accounted for. During execution, the appropriate path is followed depending upon the result of the decision. 5. Flow lines. Flow lines with arrow heads are used to indicate the flow of operation, that is, the exact sequence in which the instructions are to be executed. The normal flow of flowchart is from top to bottom and left to right. Arrow heads are required only when the normal top to bottom flow is not to be followed. However, as a good practice and in order to avoid ambiguity, flow lines are usually drawn with an arrow head at the point of entry to a symbol. Good practice also dictates that flow lines should not cross each other and that such intersections should be avoided whenever possible. 6. Connectors. Whenever a flowchart becomes complex enough that the number and direction of flow lines is confusing or it spreads over more than one page, it is useful to utilize the connector symbol as a substitute for flow lines. This symbol represents an entry from, or an exit to another part of the flowchart. A connector symbol is represented by a circle and a letter or digit is placed within the circle to indicate the link. A pair of identically labeled connector symbols is commonly used to indicate a continued flow when the use of a line is confusing. So two connectors with identical labels serve the same function as a long flowline. That is, they show an exit to some other chart section, or they indicate an entry from another part of the chart. How is it possible to determine if a connector is used as an entry or an exist point? It is. very simple - if an arrow enters but does not leave a connector, it is an exit point and program control is transferred to the identically labeled connector that does have an outlet. It may be noted that connectors do not represent any operation and their use in a flowchart is only for the sake of convenience and clarity. Additional Flowchart Symbols The basic flowchart symbols discussed above are sufficient to represent any program logic. However, these symbols do not specify in particular the type of storage media to be used, the type of I/O device to be used, or the type of processing to be done. Hence additional flowchart symbols were later defined for incorporating such details in flowcharts if so desired by the programmers. Some of the most commonly used additional flowchart symbols are shown in Figure 11.3 and their functions are discussed below. 1. Printed Document. This symbol is used to show that either the input is in the form of printed document or the output is to be produced in hard copy form using a printer. 2. Display. This symbol is used to show that the output is to be displayed on a monitor. 3. Keyed Input. This symbol is used to show that the data is to be input using a keyboard device. 4. Manual Input. This symbol is used to show that the data is to be input using an input device that requires manual operation such as keyboard, voice input system, touch screen kiosk, etc. 5. Auxiliary Operation. This symbol is used to represent an operation that is performed by a machine that is not directly under the control of the computer system to be used for
the main processing function. However, it is important to perform the auxiliary operation for the successful completion oi the overall processing job. 6. Manual Operation. This symbol is used to represent any offline activity to be performed by a human being. Again it is important to perform this activity for the successful completion of the overall processing job. 7. Online Storage. This symbol is used to represent any data stored on an online storage device such as a disk. 8. Offline Storage. This symbol is used to represent any data stored on an offline device storage device suet as a tape. 9. Magnetic Tape. This symbol is used to directly represent the use of magnetic tape as a storage device for the storage of some data. 10. Magnetic Disk. This symbol is used to directly represent the use of magnetic disk as a storage device for the storage of some data. 11. Communication Link. This symbol is used to represent data transmission from one location t( another across a network. 12. Annotation. This symbol is used to provide additional explanatory notes to some parts of the flowchart to make it easier to follow and understand. The explanatory note is written inside the three sided annotation enclosure and the dotted line connecting the annotation enclosure is drawn up to that part of the flowchart to which the annotation corresponds. 13. Predefined Process. This symbol is used to represent a group of operations that form a separate module and may be detailed in a separate flowchart. 14. Merge. This symbol is used to represent a process that combines two or more sets of items into on set. 15. Extract. This symbol is used to represent a process that separates one or more specific sets of item from a given set of items. 16. Sort. This symbol is used to represent a process that arranges a given set of items into a specified sequence. 17. Collate. This symbol is used to represent a process that performs both merge and extract operations. That is, it consists of a process that combines two or more sets of items into one set and another process that separates one or more specific sets of items from the generated set. Printed document Manual Input Online Storage Magnetic Disk Display Predefined Process Sort Auxiliary Operation Offline Storage Communication Links Merge Keyed Input Manual Operation Magnetic Tape
Annotation Extract Collate Sample Flowcharts A flowchart should be drawn using the symbols mentioned above. Describing an algorithm in the form of a flowchart is not very difficult. What is required is some common sense and a little practice. The art of flowcharting is introduced below with the help of some simple examples. Example 11.3. A student appears in an examination that consists of total 10 subjects, each subject having maximum marks of 100. The roll number of the student, his/her name, and the marks obtained by him/her in various subjects are supplied as input data. Such a collection of related data items that is treated as a unit is known as a record. Draw a flowchart for the algorithm to calculate the percentage marks obtained by the student in this examination and then to print it along with his/her roll number and name. Solution: The flowchart for the algorithm of this problem is shown in Figure 11.4. The first symbol is a terminal symbol labeled "Start". It shows that this is the starting point or beginning of our flowchart logic. It does not mean that the computer is to be turned on or that anyone is to press a start button. The second symbol is an I/O symbol that is labeled specifically to show that this step is "Read input data". This step will input the roll number, name, and the marks obtained by the student from an input device into the main storage of the computer system. The third symbol is a processing symbol, which is suitably labeled to indicate that at this step, the computer will add the marks obtained by the student in various subjects and then store the sum in a memory location, which has been given the name Total. The fourth symbol is again a processing symbol. The label inside it clearly indicates that the percentage marks obtained by the student is calculated at this stage by dividing Total by 10 and the result is stored in a memory location, which has been given the name Percentage. The fifth symbol is an I/O symbol and is labeled "Write output data". This logical step in the flowchart outputs the data desired as output (roll number, name and percentage marks obtained) on an output device such as a printer. Note that details such as the roll number, name and the marks or percentage being input or output are not a part of the logical steps of data input or data output operations. This information already appears in the system design documents and will be included in the computer program as input and output descriptions. The sixth symbol is a terminal symbol labeled "Stop". This symbol indicates the conclusion of our logic, that is, the conclusion of the computer program. The various symbols used in the flowchart are connected by directed flow lines to indicate the sequence in which the instructions are to be executed. Start Read input data Add marks of all subjects giving Total Percentage = Total 110 Write output data
Stop The logic depicted in Figure 11.4, therefore, will read the student's record, calculate the percentage marks obtained by him/her, print one line of output, and then stop. One would certainly not like to use a computer to solve a trivial problem such as this. However, if we have to compute the percentage marks obtained by several students in the same examination then we may like to take the help of a computer. The next example illustrates how to do this. Example 11.4. 50 students of a class appear in the examination of Example 11.3. Draw a flowchart for the algorithm to calculate and print the percentage marks obtained by each student along with his/her roll number and name. Solution: Since all the students have appeared in the same examination, so the process of calculation and printing the percentage marks obtained by each student will basically remain the same. The same process of reading the input data, adding the marks of all subjects, calculating the percentage, and then writing the output data has to be repeated for all the 50 students. Hence, an easy solution that comes to ones mind for this problem is to repeat the intermediate four symbols of Figure 11.4 fifty times. However if that is done, a total of 202 (50 x 4 + 2) flowchart symbols will have to be drawn. Obviously, this will be a very time consuming and tedious job and hence is not desirable. We will now see how to solve this problem in a simpler way. In a situation where the same logical steps can be repeated, the flow line symbols are used in a flowchart to indicate the repetitive nature of the logic in the form of a process loop. Figure 11.5 illustrates a flowchart with a process loop. Note the arrow head on the flow line that forms the loop. It points upward indicating that as soon as the "Write" operation is over, the control will flow back to the "Read" operation. Thus, the process loop of Figure 11.5 solves the problem of an exceedingly long flowchart by reusing the same logical steps over and over again. However, the flowchart of Figure 11.5 is incomplete because the process loop has introduced a new problem. The process loop shown does not have a logical ending. It will continue to attempt to repeat those four steps until someone manually cancels the job. This is an example of an infinite loop and hence the flowchart of Figure 11.5 does not represent an algorithm because an algorithm must terminate. So we have to find out a way of terminating the algorithm. This is done by the use of a decision symbol. Figure 11.6 shows a flowchart, which uses a decision step to terminate the algorithm. In this flowchart, another variable Count has been introduced which is initialized to zero outside the process loop and is incremented by 1 after processing the data for each student. Thus, the value of Count will always be equal to the number of students whose data has already been processed. At the decision step, the value of Count is compared with 50, which is the total number of students who have appeared for the examination. The steps within the process loop are repeated until the value of Count becomes equal to
50, the instruction at the decision step causes the control to flow out of the loop and the processing stops because a terminal symbol labeled Stop is encountered. Thus, the flowchart of Figure 11.6 is a complete and correct solution to the problem of Example 11.4. Although the flowchart of Figure 11.6 is a correct solution to the given problem, it suffers from two major drawbacks. The first drawback is that in order to make the decision at the decision step, one must know the exact number of students who appeared in the examination. Suppose the examination of Example 11.3 is a university examination in which the total number of students who appeared for the examination is too large (say more than ten thousand). In such a situation, the counting of the total number of input records (data for each student) becomes a tedious job. Even then, if we do stick on counting the input records manually and supply the number of records to be compared against Count in the decision box and suppose while counting we make some error then the logic will not work. If the supplied number is less than the actual number of input records, then the computer will not process the data for last few students. And if the supplied number is more than the actual number of input records, the computer will try to read more records than what is supplied which will cause an error in the logic. Start Read input data Add marks of all subjects giving Total Percentage = Total 110 Write output data Read input data The second drawback is that the flowchart of Figure 11.6 is not a generalized solution to the given problem. Suppose the examination is conduced every year and so we will like to use the same program every year to process the students' data. However, the number of students appearing in the examination may not remain the same every year. This year it may be 50, but next year it can be 55 or 60 or anything. So if the computer program to solve this problem was based on the flowchart of Figure 11.6, the statement concerned with the decision step in that program will have to be changed again and again to supply the exact number of students. This is not a good programming practice. We saw that a good algorithm (program) should be general in nature. For example, in this case we should write a program that need not be modified every time even if the total number of students changes. The flowchart of Figure 11.6 suffers from these drawbacks because in this flowchart the process loop is being controlled by counting. When the loop has been executed 50 times, the decision will cause execution to proceed to the "Stop", thus terminating processing. (The reader should carefully step through the logic to ensure that the loop will be executed exactly 50 times and not 49 or 51 times). To avoid these drawbacks, another method can be adopted to control the process loop. In this method, the end of input data is marked by a trailer record that is the last data record in the input is followed by a record whose sole purpose is to indicate that the end of the input data has been reached. Suppose the first 7 characters of the input record represents a student's roll number
(Rollno). Since 0000000 is never used as a roll number, a value of 0000000 as the first 7 characters can be used to represent the trailer record. As each input record is processed, the Rollno can be compared with 0000000 to determine if processing is complete. The logic of this process is illustrated in the flowchart of Figure 11.7. It is important to recognize that the programmer would have to include instructions in the program, which specifically compare the Rollno to 0000000 in order to determine whether to continue or to terminate. Start Count = 0 Read input data Add marks of all subjects giving Total Percentage = Total I 10 Write output data Add1 to Count Is count = 50? Yes Stop If no Read input data The concept of a trailer record centers around the notion of selecting a field (a particular item of data) in the input record which will be used to indicate the end of data and then selecting a trailer value also known as sentinel value which will never occur as normal data value for that field. The roll number of 0000000 is a good example. It may also be noted that when a trailer record is used to mark the end of input data, the decision box used for checking the trailer value should almost always be flowcharted immediately after the input symbol. Start Read input data Is Rollno = 0000000? If yes than Stop If no Add marks of all subjects giving Total Percentage = Total/10 Write output data Read input data Example 11.5. For the examination of Example 11.3, we want to make a list of only those students who have passed obtained 30% or more marks) in the examination. In the end, we also want to print out the total number of students who have passed. Assuming that the input data of all the students is terminated by a trailer record that has sentinel value of 9999999 for Rollno, draw a flowchart for the algorithm to do the above job.
Solution: The flowchart in Figure 11.8 is a solution to this problem. There are two decision symbols in this flowchart. The first decision symbol checks for a trailer record by comparing Rollno against the value 9999999 to determine if processing is complete. The second decision symbol is used to check whether the student has passed or failed by comparing the percentage marks obtained by him/her against 30. If the student's Percentage is equal to or more than 30 then he/she has passed otherwise failed. Note from the flowchart that the operation "Write output data" is performed only if the student has passed. If the student has failed, we directly perform the operation "Read input data" without performing the "Write" operation. This ensures that the output list provided by the computer will contain the details of only those students who have passed in the examination. Start Count= 0 Read input data Is Rollno = 9999999? If yes Write count Stop If no Add marks of all subjects giving Total Percentage = Total/10 Is Percentage = > 30? Yes Write output data Add1 to count If No Read input data Another point to be noted in this flowchart is the use of the variable Count. This variable has been initialized to zero in the beginning and is incremented by 1 every time the operation "Write output data" is performed. But we have seen that the operation "Write output data" is performed only for the students who have passed. Hence, the variable Count will be incremented by 1 only in case of students who have passed. Thus, the value of Count will always be equal to the number of students whose data has already been processed and who have been identified as passed. Finally, when the trailer record is detected, the operation "Write Count" will print out the final value of Count that will be equal to the total number of students who have passed the examination. Example 11.6. Suppose the input data of each student for the examination of Example 11.3 also contains information regarding the sex of the candidate in the field named Sexcode that can have values M (for male) or F (for female). We want to make a list of only those female students who have passed in second division (obtained 45% or more but less than 60% marks). In the end we also want to print out the total number of such students. Assuming
that the input data of all the students is terminated by a trailer record that has a sentinel value of Z for Sexcode, draw a flowchart for the algorithm to do the above job. Solution: The flowchart in Figure 11.9 is a solution to this problem. There are four decision symbols in this flowchart. The first one checks for a trailer record by comparing Sexcode against the value Z to determine if processing is complete. The second one is used to check whether the candidate is female or not by comparing the Sexcode of that candidate against F. Note that if the Sexcode is not F, that is, the candidate is not a female, we do not process the data of that student and return back to perform the operation of reading input data. This step ensures that the data of only female students will be taken for further processing. The last two decision symbols in the flowchart are used to check whether the student has passed in second division or not. The first of these decisions is used to ensure that the student has scored 45% or more marks. If she has scored less than 45% then it means that she is not a second divisioner and hence without making any further check we return back to the operation of reading input data. In case the student has scored 45% or more marks then we go one step further in the logic and by using the fourth decision in the flowchart we check whether her marks are less than 60% by comparing the Percentage against 60. If the condition at this step turns out to be false then it means that the student has scored 60% or more marks and hence she is a first divisioner and not a second divisioner. So once again we return back to read a new data without writing any output data. If the condition inside the fourth decision symbol turns out to be true then the female candidate can be classified to be a second divisioner. Hence in this case only we perform the operation "Write output data" and subsequently increment the value of Count by 1. It is suggested that the reader should go through the logic of this flowchart again and again until he/she is convinced that the output list provided by the computer will contain the details of only those female students who have passed in second division. The reader should also get convinced that finally when the trailer record is detected, the operation "Write Count" would print out the value of Count that will be equal to the total number of female students who have passed in second division. This flowchart is an example of a multiple-decision chart. The flowchart of Figure 11.9 has been reproduced in Figure 11.10 to illustrate the use of connectors. There are four exit connectors having the label 1 all of which indicate a flow of control to the flowchart symbol having an entry connector labeled 1. This symbol is the input symbol in the flowchart. Similarly the exit connector having a label 2 indicates a flow of control to the entry connector labeled 2. The reader should compare the flowcharts of Figure 11.9 and Figure 11.10 and should get convinced that both the flowcharts represent the same logic. A flowchart may seem simple to prepare, but you will find that much practice is needed in order to think through a problem in discrete, logical steps, to assume nothing and to forget nothing. Moreover, not everyone will tackle a problem in exactly the same way and, in consequence, several different flowcharts could be drafted for the same problem. It may also be noted that a completed flowchart is not a complete computer program. It is
only an aid to programming. For a given problem, it defines the procedure and the logic involved. From the examples that have been discussed above, we are in a better position to understand what this "logic" means. Levels of Flowcharts There are no set standards on the amount of detail that should be provided in a flowchart. A flowchart that outlines the main segments of a program or that shows less detail is a macro flowchart. On the other hand, a flowchart with more detail is a micro flowchart, or detailed flowchart. For example, let us consider the examination problem that we have already discussed. In all the flowcharts of the examination problem, there is a processing box having the instruction "Add marks of all subjects giving Total. In order to display how the value of Total is computed, a detailed flowchart can be drawn as shown in Figure 11.11. In a similar manner, the I/O boxes for the "Read" and "Write" operations can also be converted to a detailed flowchart. Flowcharting Rules While programmers have a good deal of freedom in creating flowcharts, there are a number of general rules and guidelines recommended by the American National Standards Institute (ANSI) to help standardize the flowcharting process. Various computer manufacturers and data processing departments usually have simila flowcharting standards. Some of these rules and guidelines are as follows: 1. First chart the main line of logic, then incorporate detail. 2. Maintain a consistent level of detail for a given flowchart. 3. Do not chart every detail or the flowchart will only be a graphic representation, stepby-step, of thi program. A reader who is interested in greater details can refer to the program itself. 4. Words in the flowchart symbols should be common statements and easy to understand. It i recommended to use descriptive titles written in designer's own language rather than in machine oriented language. 5. Be consistent in using names and variables in the flowchart. 6. Go from left to right and top to bottom in constructing flowcharts. 7. Keep the' flowchart as simple as possible. The crossing of flow lines should be avoided as far a practicable. 8. If a new flowcharting page is needed, it is recommended that the flowchart be broken at an input o output point. Moreover properly labeled connectors should be used to link the portions of th flowchart on different pages. Advantages and Limitations of Flowcharts Advantages The following benefits may be obtained when flowcharts are used for the purpose of program planning:
1. Better Communication. The old saying that "a picture is worth a thousand words" holds true for flowcharts also. Since, a flowchart is a pictorial representation of a program, it is easier for programmer to explain the logic of a program to some other programmer or to his/her boss through flowchart rather than the program itself. 2. Effective Analysis. A macro flowchart that charts the main line of logic of a software system becomes a system model that can be broken down into detailed parts for study and further analysis c the system. 3. Effective Synthesis. A group of programmers are normally associated with the design of bi software systems. Each programmer is responsible for designing only a part of the entire system. S initially, if each programmer draws a flowchart for his/her part of design, the flowcharts of all the programmers can be placed together to visualize the overall system design. Any problem in linking the various parts of the system can be easily detected at this stage and the design can be accordingly modified. Flowcharts are thus used as working models in the design of new programs and software systems. 4. Proper Program Documentation. Program documentation involves collecting, organizing, storing, and otherwise maintaining a complete historical record of programs and the other documents associated with a system. Good documentation is needed for the following reasons: - Documented knowledge belongs to an organization and does not disappear with the departure (resignation/retirement) of a programmer. - If projects are postponed, documented work will not have to be duplicated. - If programs are modified in the future, the programmer will have a more understandable record of what was originally done. From what we have seen of the nature of flowcharts, it is obvious that they can provide valuable documentation support. 5. Efficient Coding. Once a flowchart is ready, programmers find it very easy to write the concerned program because the flowchart acts as a road map for them. It guides them to go from the starting point of the program to the final point ensuring that no steps are omitted. The ultimate result is an error free program developed at a faster rate. 6. Systematic Debugging. Even after taking full care in program design, some errors may remain in the program because the designer might have never thought about a particular case. These errors are called bugs and the process or removing these errors is known as debugging. Once a bug is detected, it is easier to find out the reason for that bug by going through the logic of the program in flowchart form. A flowchart is very helpful in detecting, locating, and removing mistakes (bugs) in a program in a systematic manner. 7. Systematic Testing. Testing is the process of confirming whether a program will successfully do all the jobs for which it has been designed under the specified constraints. For testing a program, different sets of data are fed as input to that program to test the different paths in the program logic. For example, to test the complete logic of the program for Example 11.6, the following sets of data are necessary: - Data for a male candidate. - Data for a female candidate who has scored less than 45%. - Data for a female candidate who has exactly scored 45%. - Data for a female candidate who has scored more than 45% but less than 60%. - Data for a female candidate who has exactly scored 60%.
- Data for a female candidate who has scored more than 60%. - And obviously in the end the trailer data having sentinel value. A flowchart proves to be very helpful in designing the test data for systematic testing of programs. Limitations In spite of their many obvious advantages, flowcharts have several limitations, which are as follows: 1. Flowcharts are very time consuming and laborious to draw with proper symbols and spacing especially for large complex programs. In this chapter, you have seen examples of small program flowcharts developed for relatively small programs. You can very well imagine how difficult would be to develop a detailed program flowchart for a program containing over 50,000 statements. 2. Owing to the symbol-string nature of flowcharting, any changes or modifications in the program logi will usually require a completely new flowchart. Redrawing a flowchart being a tedious task, man programmers do not redraw or modify the corresponding flowchart when they modify their program: This leaves the program and its flowchart in an inconsistent state. That is, the logic used in the program and that shown in its flowchart do not match. This defeats the purpose of use of flowchart as documentation support for programs. To take care of this problem, many companies use software tools that automatically generate flowcharts directly from the program code. These software tools read the program's instructions and draw a flowchart of its logic. That is, this is a backward approach in which flowcharts are designed from program codes mainly for documentation purpose. 3. There are no standards determining the amount of detail that should be included in a flowchart. Because of such limitations, many organizations are now reducing the amount of flowcharting used. In its place, they are using alternative tools for program analysis, two of which are briefly discussed below. DECISION TABLES What is a Decision Table? Decision tables are used to define clearly and concisely the word statement of a problem in a tabular form. The can prove to be a powerful tool for defining complex program logic. As the name implies, decision tables reveal what decisions or actions the computer is to take as a result of the input data. When the computer has to make large number of decisions or if there are a large number of different branches within a program, decision tables a particularly useful. In these cases, decision tables are preferred to flowcharts. Most decision tables have six parts. The basic format of a decision table is shown in Figure 11.12. The first p< of a decision table is "Table Heading" that contains the name and/or number of the table. For some large applications, two or more decision tables may be used in the analysis part of program development. The second part of the table, known
as "Condition Stub", describes the conditions that could exist in the program logic. Thus the contents of condition stub correspond to the conditions contained in the decision symbols of a flowchart "Action stub", the third part of the decision table, contains the action statements. These statements correspond to the statements located in non-decision symbols of a flowchart. While the condition statements reveal the possible states of the input data, the action statements describe the possible actions of the computer system. The right hand side of the decision table contains the rule numbers (part 4), the actual conditions (part 5), and the actions taken by the computer (part 6). The condition entries correspond to the paths leading out from decision symbols. Table Heading
Decision Rules
Condition Stub
Condition Entries
Action Stub
Action Entries
Figure 11.12. Format of a decision table. The steps to be followed for constructing a decision table are as follows: 1. Properly define the problem that has to be solved by computer. 2. List out all the conditions to be tested in the problem. 3. List out the corresponding actions that should be taken with each combination of conditions. 4. Form a decision table using the two lists. Sample Decision Table A decision table for the problem of Example 11.6 that was charted in Figure 11.9 is shown in Figure 11.13. In this table, each rule number is a given condition followed by a specific action to be taken by the computer. The six rule numbers, the six actual conditions, and the associated actions taken by the computer system are discussed below. Examination Problem of Example 11.6 Condition Statements Sexcode = Z Sexcode = F Percentage = > 45 Percentage < 60 Action Taken Calculate Percentage
Decision Rule Number . 1 2 1 3 4 | 576 N Y Y Y
N Y Y N
N Y N -
N Y X
N N -
Y -
Write output data X Add 1 to Count X Read next student's record X X X X Write Count X Stop X Figure 11.13. Decision table for the examination problem of Example 11.6. Rule 1: The student is a female, and the percentage marks obtained is 45% or more, and the percentage marks obtained is less than 60%. The computer should write the output data, add 1 to Count, and then read the next student's record. It is a case of a female student who has passed in second division. Rule 2: The student is female, and the percentage marks obtained is 45% or more, and the percentage marks obtained is not less than 60%. The computer should directly read the next student's record without performing any other operation. It is the case of a female student who has scored 60% or more marks and hence is a first divisioner and not a second divisioner. Rule 3: The student is a female and the percentage marks obtained is not 45% or more. A dash (-) in this column against the last condition means that this condition is irrelevant in arriving at the action. In fact, if Percentage is not equal to or greater than 45, then it has to be less than 60. Percentage greater than or equal to 60 is logically impossible. Such conditions in the decision table, which are irrelevant in arriving at an action, are known as don't care conditions and are denoted by a dash. For this rule, the computer should directly read the next student's record without performing any other operation. It is the case of a female student who has scored less than 45% and hence she is not a second divisioner. Rule 4: The student is a female, and the other conditions are don't care conditions. The computer should calculate Percentage after adding marks of all subjects and then proceed to test further conditions. It is the case of a female student whose Percentage is yet to be calculated. Rule 5: The student is a male, and the other conditions are don't care conditions. The computer should directly read the next student's record without performing any other operation. It is the case of a male student. Rule 6: In all the previous five rules, it was ascertained that the current record is not a trailer record because all these rules have a value N (No) for the condition Sexcode = Z. In this rule, the Sexcode is equal to Z, which indicates a trailer record, and hence the computer should write the value of Count and then "Stop". The other conditions in this rule are don't care conditions. It is the case of a trailer record. In the decision table, "Y" means yes, "N" means no, "-" means don't care, and "X" means the computer should take this action.
Advantages and Limitations of Decision Tables Advantages Decision tables are normally used in place of flowcharts because of the following reasons: 1. They provide concise descriptions of logically complex situations. 2. They are easier to draw and change than flowcharts. 3. They provide more compact documentation. A small table can replace several pages of flowcharts. 4. It is also easier to follow a particular path down one column than through several flowchart pages. Limitations Decision tables, however, are not very popular and are not so widely used as flowcharts because: 1. Large decision tables can become incomprehensible and difficult to modify. 2. Flowcharts are better able to express the total sequence of events needed to solve a problem. 3. Flowcharts are more familiar to, and are preferred by, many programmers and beginners. PSEUDOCODE What is Pseudocode? Pseudocode is another programming analysis tool that is used for planning program logic. "Pseudo" means imitation or false and "Code" refers to the instructions written in a programming language. Pseudocode, therefore, is an imitation of actual computer instructions. These pseudo-instructions are phrases written in ordinary natural language (e.g., English, French, German, etc.) and hence cannot be understood by the computer. Instead of using symbols to describe the logic steps of a program, as in flowcharting, pseudocode uses a structure that resembles computer instructions. When pseudocode is used for program planning, a programmer can concentrate solely on developing the logic of the program without worrying about the syntax for writing the program instructions because pseudocode does not have any syntax rules for formulating instructions. Once the programmer is convinced that the program logic is sound, he/she can easily convert the pseudocode into a suitable programming language that can be run on a computer. Because it emphasizes the design of the program, pseudocode is also called Program Design Language (PDL). Pseudocodes for Basic Logic (Control) Structures During the early days of program development, many programmers developed program logics for large programs with mazes of branches (jumps from one portion of the program to another) that altered the sequence of processing operations. Such programs are now referred to as "spaghetti code" because their program flowcharts appeared more like a
plate of spaghetti than like logical analyses of programming problems. Understanding the logic of such programs was very difficult for someone other than the developer and many times even for the developer after a lapse of few months. Hence these programs were very difficult to modify for incorporating suggested enhancements and quickly became nightmares for those responsible for their maintenance. This problem was soon identified as a major research area by computer scientists. They started looking into what factors contributed to unnecessary program complexity and how to develop program logics that are easier to comprehend resulting in easier maintenance of the corresponding programs. As a result of the research efforts made in this area, it was discovered that any program logic, no matter how complex, could be expressed by using only the following three simple logic (control) structures: 1. Sequence logic, 2. Selection logic, and 3. Iteration (or looping) logic. It was found that when programs are structured by using only these three logic structures, they can be read from top to bottom and are easier to understand. That is, by conceptualizing the logic of a program in the form of these three logic structures, programmers can avoid writing spaghetti code and produce programs that are easy to understand and maintain. It was found that this also helped in reducing program errors and the time spent in program testing. The use of these three basic control structures resulted in a more scientific approach to solving a programming problem and was later termed as structured programming technique. After realizing the advantages of structured programming technique, many organizations now impose the use of the three basic logic structures in all their programs. It is generally advised as good programming practices to develop program logics and write programs using the three basic logic structures. Hence in the discussion below we will see the pseudocodes for the three basic logic structures (flowcharts of these logic structures have also been shown for those who are rhore comfortable with flowcharts). Sequence Logic Sequence logic is used for performing instructions one after another in sequence. Thus, for sequence logic, pseudocode instructions are written in the order, or sequence, in which they are to be performed. The logic flow of psuedocode is from the top to the bottom. Figure 11.14 shows an example of sequence logic structure. Process l Process 2 (a) Flowchart
Process 1 Process 2 (b) Pseudocode Figure 11.14. Flowchart and pseudocode for sequence structure. Selection Logic Selection logic, also known as decision logic, is used for making decisions. It is used for selecting the proper path out of the two or more alternative paths in the program logic. Selection logic is depicted as an IF...THEN...ELSE or an IF...THEN or a CASE structure. The flowcharts of Figures 11.15, 11.16 and 11.17 respectively illustrate the logic of these structures. Their corresponding pseudocodes are also given in these figures. The IF... THEN... ELSE construct says that if the condition is true, then do Process 1, else (if the condition is not true) do Process 2. Thus, in this case either Process 1 or Process 2 will be executed depending on whether the specified condition is true or false. However, if we do not want to choose between the two processes and if we simply want to decide if a process is to be performed or not, then the IF...THEN structure is used. The IF... THEN structure says that if the condition is true, then do Process 1 and if it is not true then skip over Process 1. The CASE structure is a multiple-way selection logic structure that allows to choose between more than two control paths in a program logic. It allows the selection of one of any number of statements or statement groups. It is, therefore, a generalization of the one-way (IF...THEN) and two-way (IF...THEN...ELSE) selection logic structures. In fact, one-way and two-way selection logic structures can be built with a multiple-way selection logic structure. The CASE statement says that if the value of Type is equal to Type-1 then execute Process 1, if it is equal to Type-2 then execute Process 2, if it is equal to Type-3 then execute Process 3, and so on. IF Condition Yes THEN Process 1 ELSE Process 2 ENDIF (a) Flowchart In all these structures, Process 1, Process 2, Process 3, etc. can actually be one or more processes. They are not limited to a single process. ENDIF is used to indicate the end of the IF,... THEN and IF... THEN.. .ELSE decision structures and ENDCASE is used to indicate the end of the CASE structure. Iteration (or Looping) Logic Iteration logic is used to produce loops in program logic when one or more instructions may be executed several times depending on some condition. It uses two structures called the DO...WHILE and the REPEAT...UNTIL. They are illustrated by flowcharts in Figure
11.18 and Figure 11.19 respectively. Their corresponding pseudocodes are also given in these figures. Both DO...WHILE and REPEAT...UNTIL are used for looping. In case of DO...WHILE, the looping will continue as long as the condition is true. The looping stops when the condition is not true. On the other hand, in case of REPEAT... UNITL, the looping continues until the condition becomes true. That is, the execution of the statements within the loop is repeated as long as the condition is not true. Notice that in case of DO... WHILE, the condition for looping is checked at the beginning of the loop. Hence if the condition is found to be false when the loop is first encountered, the processes in the loop will not be executed at all (not even once). On the other hand, in case of REPEAT... UNTIL, the condition for looping is checked at the end of the loop. Hence the processes inside the loop will be executed at least once. In both the DO...WHILE and REPEAT...UNITL, the loop must contain a statement that will change the condition that controls the loop. If it does not, the looping will continue without end, which is the case of an infinite loop. Remember that no program should contain an infinite loop. Also note that the condition is tested at the top of the loop in the DO...WHILE and at the bottom of the loop in the REPEAT... UNTIL. ENDDO marks the end of a DO...WHILE structure and UNTIL followed by some condition marks the end of the REPEAT...UNTIL structure. Condition? True Process1 Process n If false (a) Flow chart DO WHILE Condition Process 1 Process n ENDDO (b) Pseudocode Figure 11.18. Flowchart and pseudocode for DO...WHILE iteration structure. Sample Pseudocode The pseudocode version of the logic of the problem of Example 11.6 that was charted in Figure 11.9 is shown in Figure 11.20. In the pseudocode example, the first line initializes the value of Count to zero and the second line reads the input data of the first student. The third line is the beginning of a loop using the DO.. .WHILE structure. It indicates that the loop will continue so long as the trailer Sexcode is not equal to Z, that is, as long as the trailer record is not fond. In this example, a series of decisions followed by an instruction to read next student's record are included within the loop. The first statement within the DO.. .WHILE loop asks, "Is the Sexcode equal to F?" If the answer is yes,
Pergentage is calculated and again the third statement within the loop asks, "Is Percentage equal to or greater than 45?" If it is, then "Is Percentage less than 60?" This is a series of three IF...THEN decision structures. Each one ends with an ENDIF vertically aligned below the appropriate IF. The two instructions - "Write output data" and "Add 1 to Count" are performed only if all the three conditions (that of Sexcode being F, Percentage being equal to or more than 45, and Percentage being less than 60) are found to be true (answered yes). If any of the three conditions is not true, the logic path goes to the statement that reads the next student's record. After the last student's record is processed, the trailer record for which the value of Sexcode is Z is encountered. This will cause the DO...WHILE loop to stop, because the condition (Sexcode not equal to Z) is no longer true. When the DO.. .WHILE condition is no longer true, the next logical step will be the instruction following the ENDDO. At this stage, the value of Count will be printed (Write Count) and finally the program execution will stop (Stop). Set Count to zero Read first student record DO WHILE Sexcode is not equal to Z IF Sexcode = F THEN Calculate Percentage IF Percentage = > 45 THEN IF Percentage < 60 THEN Write output data Add 1 to Count ENDIF ENDIF ENDIF Read next student record ENDDO Write Count Stop One important feature of pseudocode as a programming tool is the use of indentation. Each statement within the DO...WHILE loop is indented, to show that it is part of the loop. Similarly, the statements within each IF...THEN structure is indented properly to clearly specify the statements, which are part of each structure. The use of indentation in pseudocode is the same technique used with the various programming languages. Its sole purpose is to clarify the logical structure of the program. With this technique, we can tell at a glance, which statements make up each of the logic structure of the total program logic. To fully appreciate this factor, the reader should compare the equivalent nonindented pseudocode of Figure 11.21 with that of Figure 11.20. The difference in clarity would be far greater if this were a longer pseudocode covering, for instance, one or more pages. Set Count to zero
Read first student record DO WHILE Sexcode is not equal to Z IF Sexcode = F THEN Calculate Percentage IF Percentage = > 45 THEN IF Percentage < 60 THEN Write output data >• Add 1 to Count ENDIF ENDIF ENDIF Read next student record ENDDO Write Count Stop Advantages and Limitations of Pseudocode Advantages Pseudocode has three main advantages: 1. Converting a pseudocode to a programming language is much more easier than converting a flowchart or a decision table to a programming language. 2. As compared to a flowchart, it is easier to modify the pseudocode of a program logic when program modifications are necessary. 3. Writing of pseudocode involves much less time and effort than drawing an equivalent flowchart. Pseudocode is easier to write than an actual programming language because it has only a few rules to follow, allowing the programmer to concentrate on the logic of the program. Limitations Pseudocode, however, suffers from the following limitations: 1. In case of pseudocode, a graphic representation of program logic is not available. 2. There are no standard rules to follow in using pseudocode. Different programmers use their own style of writing pseudocode and hence communication problem occurs due to lack of standardization. 3. For a beginner, it is more difficult to follow the logic of or write pseudocode, as compared to flowcharting. Points to Remember 1. In order to ensure that the program instructions are appropriate for the problem at hand and in the correct sequence, programs must be planned before they are written. 2. The term algorithm refers to the logic of a program. It is a step-by-step description of how to arrive at a solution of the given problem. 3. The primary factors that are often used to judge the quality of an algorithm are the time needed to execute it, the memory space required to execute it, the accuracy of the
results produced by it, and its capability to handle a range of input data instead of a single input data (often referred to as generality property). 4. When an algorithm is represented in the form of a programming language, it becomes a program. Thus any program is an algorithm although the reverse is not true. 5. Besides represented as programs, algorithms are often represented as flowcharts, pseudocodes and decision tables. 6. A flowchart is a pictorial representation of an algorithm. It is often used by programmers as a program planning tool for organizing a sequence of steps necessary to solve a problem by a computer. 7. Decision tables are used to define clearly and concisely the word statement of a problem in a tabular form. They can prove to be a powerful tool for defining complex program logic. 8. Pseudocode is a programming analysis tool that allows programmers to plan program logic by writing the program instructions in an ordinary natural language such as English. 9. Structured Programming is a more scientific approach to solving a programming problem by using only the three basic logic (control) structures namely, sequence logic, selection logic an iteration (or looping) logic. 10. Structured programming helps in producing programs that are easy to understand and maintain. It also helps in reducing program errors and the time spent in program testing. 11. Sequence logic is used for performing instructions one after another in sequence. 12. Selection logic is used for selecting the proper path out of the two or more alternative paths in the progran logic. It uses three control structures called IF.. .THEN, IF.. .THEN.. .ELSE and CASE. 13. Iteration (or looping) logic is used for producing loops in program logic when one or more instructions ma; be executed several times depending on some condition. It uses two control structures called DO...WHIL1 and REPEAT...UNTIL. Questions 1. Why is it advisable to plan the logic of a program before writing it? 2. What is an algorithm? What are the characteristics necessary for a sequence of instructions to qualify as a algorithm? 3. How do we normally judge the quality of an algorithm? 4. What are the commonly used ways to represent algorithms? Which of these can be used for solving the corresponding problem on a computer? 5. Any program is an algorithm although the reverse is not true. Discuss this statement. 6. What is a program planning tool? How is it useful? Name two commonly used program planning tools. 7. What is a flowchart? 8. How does a flowchart help a programmer in program development? 9. Can a flowchart be drawn for a task if the person drawing the flowchart does not know how to perform the task manually? Discuss. 10. What are the various basic symbols used in flowcharting? Give their pictorial representation. 11. Describe the function of the various basic flowcharting symbols.
12. Why are there standards for the symbols used in drawing flowcharts? 13. What is a record? A trailer record? 14. What is a sentinel value? Discuss its use. 15. What is a process loop? An infinite loop? 16. Why is it necessary to avoid infinite loops in program design? 17. "A loop consists of a body, a test for exit condition, and a return provision." Discuss this statement. 18. What is a generalized algorithm? Why should programs be general in nature? 19. Discuss the difference between loop control by counting and loop control by the use of sentinel value. Which is preferable and why? 20. How can a counter be used to keep track of the number of times a loop has been executed? 21. Is it possible to have more than one flowchart for a given problem? Give reasons for your answer. 22. What is the difference between a macroflowchart and a microflowchart? Illustrate with an example. 23. What are the various guidelines to be followed while drawing a flowchart? 24. Discuss the advantages and limitations of flowcharting. 25. Why is proper documentation of a program necessary? 26. What are program bugs? What is debugging? 27. What is meant by testing a program? How is it done? 28. What are decision tables? When are they preferred to flowcharts? 29. What are the various steps to be followed for constructing a decision table? 30. Draw the basic format of a decision table and discuss the role of each part. 31. What are don't care conditions in decision tables? 32. List out the various advantages and limitations of decision tables. 33. What is a pseudocode? Why is it so called? Give another name for pseudocode. 34. What is a spaghetti code? Why is it so called? What are its major drawbacks? 35. What are the three basic logic structures used in writing structured programs? Discuss the use of each. 36. What is structured programming? What are its main advantages? 37. Give an example to illustrate the use of each of the three basic logic structures used in structun programming. 38. Draw flowcharts for the two different structures used for selection logic. 39. What is the difference between the IF.. .THEN and the IF.. .THEN.. .ELSE structures? 40. Draw flowcharts for the two different structures used for iteration logic. 41. Both DO...WHILE and REPEAT...UNTIL are used for looping. Discuss the difference between the tv structures. 42. If the condition for looping is found to be false when the loop is encountered for the first time, how mai times will the instructions inside the loop be executed in case of (a) DO...WHILE loop (b) REPEAT...UNTIL loop Give reasons for your answer. 43. What is the purpose of ENDIF and ENDDO statements? 44. What is indentation? Why is it used in writing pseudocodes? 45. Discuss the advantages and limitations of pseudocode.
46. Three numbers, denoted by the variables A, B, and C are supplied as input data. Draw a flowchart for t logic to pick and print the largest of the three numbers. 47. Draw a flowchart of the logical steps needed to produce a printed listing of all students over the age of 20 a class. The input records contain the name and age of the students. Assume a sentinel value of 99 for 1 age field of the trailer record. 48. Draw a flowchart of the logical steps needed to print the name and age of the oldest and the youngest student in a class. The input records contain the name and age of the students. Assume a sentinel value of 99 for the age field of the trailer record. 49. The first 20 records in a data set are to be read and printed. Draw a flowchart for the algorithm to do this job. Make sure that the processing stops after the twentieth record. 50. Input data regarding the information of employees of a company has been supplied. The first field of each input record contains the employee number (EmployeeNo). Assume that the input data of all the employees is terminated by a trailer record having a sentinel value of 99999 for EmployeeNo. Draw a flowchart for the logic to count and print the total number of input records, that is, the total number of employees. 51. For the employees problem of Question 50, we want to count and print the number of only male employees in the age group of 25 to 30. Assume that the input records contain SexCode and Age fields to provide this information. Draw a flowchart for the algorithm to perform this job. 52. Suppose that a population survey has been carried out in a given city, and that the information received from the survey has been stored in a computer, one record per citizen. That is, each record contains the name, address, sex, age, profession, etc., of one citizen. We want to print the details of all the adults (aged 18 years or more) in the city under survey. Finally, we also want to print the total number of adults. Assume a suitable sentinel value for any field in the trailer record and draw a flowchart for the algorithm to do this task. 53. A set of examination papers, which have been graded with scores from 0 to 100, is to be searched to find how many of them are above 90. The total has to be printed. Prepare a flowchart to do this job. Assume a suitable sentinel value for the trailer record. 54. Each paper in a set of examination papers includes a grade of A, B, C, D, or E. A count is to be made of how many papers have the grade of A and how many have the grade of E. The total count of both types have to be printed at the end. Prepare a flowchart to perform this task. Assume a suitable sentinel value for the trailer record. 55. A shopkeeper wants to have a general program for his personal computer that will prepare bills for each customer as and when he sells goods to them. His idea is that as soon as the customer purchases some goods from his shop, he will supply the description, unit price, and the quantity purchased for each item, as input to the computer. He wants that with this information, the computer should print each item along with its unit price, quantity purchased and the total price. Finally, the computer should also print the total cost of all the items purchased by the customer. Assuming that a sentinel value of zero is used for the quantity purchased field in the trailer record, draw a flowchart for the logic to do this job. 56. An employer plans to pay bonus to each employee. Those earning Rs. 2000 or above are to be paid 10 percent of their salary; those earning less than Rs. 2000 are to paid Rs. 200. The input records contain the employee number, name, and salary of the employees.
The output to be printed should contain the employee number, name, and the amount of bonus to be paid to each employee. Draw a flowchart for the algorithm to do this job. Assume a suitable sentinel value for any of the fields of the trailer record. 57. Each employee pay record includes the hours worked and the pay rate. The gross pay is to be determined as hours worked times pay rate and is to be printed for each employee. For all hours worked in excess of 40, the overtime rate, which is 1.5 times the regular rate, is to be paid. Draw a flowchart for the problem logic to do this. Assume a suitable sentinel value for any of the input fields of the trailer record. 58. A data file contains a set of examination scores and is followed by a trailer record with a value of-1. Draw a flowchart for the logic to calculate and print the average of the scores. 59. The data file of Question 58 is expanded to include several sets of data, each requiring calculation of average. Each data set is followed by a trailer record with a value of -1; however, the last data set followed by a trailer record with a value of-2. Draw a flowchart for the logic to perform this task. 60. Suppose five numbers denoted by the variables A, B, C, D, and E are supplied as input. Draw a flowchart for the logic to print these numbers in descending order of magnitude. 61. Draw a flowchart to add up all the even numbers between 0 and 100. Before ending, print the result of 1 calculation. 62. Draw a flowchart for the logic to find out whether a given triangle ABC is isosceles. Assume that the angles of the triangle are supplied as input. Print the answer as yes or no. 63. Draw a flowchart for the logic to find out whether a given triangle ABC is a right angled triangle. Assume that the sides of the triangle are supplied as input data. Print the answer as yes or no. 64. Draw a flowchart for the logic to find out whether a given quadrilateral ABCD is a rectangle. Assume that all the four angles and four sides of the quadrilateral are supplied as input data. Print the answer as yes or no. 65. Draw a flowchart for the logic to convert a number from base 10 to a new base using the division-remainder technique. 66. Draw a flowchart for the logic to convert a number to decimal from another base. 67. Construct a decision table to solve the problem described in Question 46. 68. Construct a decision table to solve the problem described in Question 51. 69. Construct a decision table to solve the problem described in Question 54. 70. Construct a decision table to solve the problem described in Question 62. 71. Construct a decision table to solve the problem described in Question 63. 72. Write the pseudocode to solve the problem described in Question 47. . 73. Write the pseudocode to solve the problem described in Question 48. 74. Write the pseudocode to solve the problem described in Question 51. 75. Write the pseudocode to solve the problem described in Question 54. 76. Write the pseudocode to solve the problem described in Question 55. 77. Write the pseudocode to solve the problem described in Question 56. 78. Write the pseudocode to solve the problem described in Question 57. 79. Write the pseudocode to solve the problem described in Question 61. 80. Write the pseudocode to solve the problem described in Question 62.
Chapter 12 Computer Languages This chapter continues with the development of computer programs that was begun in Chapter 11. Once the planning for a computer program has been done, the next step in its development is to write the specific steps for solving the problem at hand in a language and form acceptable to a computer system. A language that is acceptable to a computer system is called a computer language or programming language, and the process of writing instructions in such a language for an already planned program is called programming or coding. The goal of this chapter is to introduce some of the common computer languages used in writing computer programs. ANALOGY WITH NATURAL LANGUAGES A language is a means of communication. We use a natural language such as English to communicate our ideas and emotions to others. Similarly, a computer language is a means of communication used to communicate between people and the computer. With the help of a computer language, a programmer tells a computer what he/she wants it to do. All natural languages (English, French, German, etc.) use a standard set of words and symbols (+, -, :,;, <, >, etc.) for the purpose of communication. These words and symbols are understood by everyone using that language. We normally call the set of words allowed in a language, the vocabulary of that particular language. For example, the words we use in English form the vocabulary of English language. Each word has a definite meaning and can be looked up in a dictionary. In a similar manner, all computer languages have a vocabulary of their own. Each word of the vocabulary has a definite unambiguous meaning, which can be looked up in the manual meant for that language. The main difference between a natural language and a computer language is that natural languages have a large vocabulary but most computer languages use a very limited or restricted vocabulary. This is mainly because a programming language by its very nature and purpose does not need to say too much. Each and every problem to be solved by a computer has to be broken down into discrete (simple and separate), logical steps which basically comprise of four fundamental operations - input and output operations arithmetic operations, movement of information within the CPU and memory, and logical or comparison operations. Each natural language has a systematic method of using symbols of that language which is defined by the grammar rules of the language. These rules tell us how to use the different words and symbols allowed in the language. Similarly, the words and symbols of a particular computer language must also be used as per set rules,
which are known as the syntax rules of the language. In case of a natural language, people can use poor or incorrect vocabulary and grammar and still make themselves understood. However, in the case of a computer language, we must stick to the exact syntax rules of the language if we want to be understood correctly by the computer. As yet, no computer "is capable of correcting and deducing meaning from incorrect instructions. Computer languages are smaller and simpler than natural languages but they have to be used with great precision. Unless a programmer adheres exactly to the syntax rules of a programming language, even down to the punctuation mark, his/her instructions will not be understood by the computer. Programming languages have improved over the years, just as computer hardware has improved. They have progressed from machine-oriented languages that use strings of binary Is and Os to problem-oriented languages that use common mathematical and/or English terms. However, all computer languages can be broadly classified into the following three categories: 1. Machine Language 2. Assembly Language 3. High-level Language We shall now examine the nature and characteristics of these three types of languages. MACHINE LANGUAGE Although computers can be programmed to understand many different computer languages, there is only one language understood by the computer without using a translation program. This language is called the machine language of the computer. The machine language of a computer is normally written as strings of binary Is and Os. The circuitry of a computer is wired in such a way that it immediately recognizes the machine language instructions and converts them into the electrical signals needed to execute them. A machine language instruction normally has a two-part format as shown in Figure 12.1. The first part of an instruction is the operation code that tells the computer what function to perform, and the second part is the operand that tells the computer where to find or store the data or other instructions that are to be manipulated. Thus, each instruction tells the computer what operation to perform and the length and locations of the data fields that are involved in the operation. Every computer has a set of operation codes called its instruction set. Each operation code (or opcode) in the instruction set is meant for performing a specific basic operation or function. Typical operations included in the instruction set of a computer are as follows: 1. Arithmetic operations
2. Logical operations 3. Branch operations (either conditional or unconditional) for transfer of control to the address given in the operand field 4. Data movement operations for moving data between memory locations and registers 5. Data movement operations for moving data from or to one of the computer's input/output devices. Figure 12.1 shows atypical single-address machine language instruction. Although some computers are designed to use only single-address instructions, many computers are designed to use multiple-address instructions that include the addresses of two or more operands. For example, the augend and addend may be the two operands of an addition operation. OPCODE (operation code)
OPERAND (Address/Location)
Figure 12.1. Instruction format. We already know that all computers use binary digits (Os and Is) for performing internal operations. Hence most computers' machine language instructions consist of strings of binary numbers. For example, a typical program instruction to print out a number on the printer might be 101100111111010011101100 The program to add two numbers in memory and print the result might look something like the following: 001000000000001100111001 001100000000010000100001 011000000000011100101110 101000111111011100101110 000000000000000000000000 This is obviously not a very easy to use language, partly because it is difficult to read and understand and partly because it is written in a number system with which we are not familiar. But it will be surprising to note that some of the first programmers, who worked with the first few computers, actually wrote their programs in binary form as above. Since human programmers are more familiar with the decimal number system, most of them will prefer to write the computer instructions in decimal, and leave the input device to convert these to binary. In fact, without too much effort, a computer can be wired so that instead of using long strings of 1 s and 0s we can use the more familiar decimal numbers. With this change, the preceding program appears as follows:
10001471 14002041 30003456 50773456 00000000 This set of instructions, whether in binary or decimal, which can be directly understood by a computer without the help of a translating program, is called a machine code or machine language program. Thus, a machine language program need not necessarily be coded as strings of binary digits (Is and Os). It can also be written using decimal digits if the circuitry of the computer being used permits this. Advantages and Limitations of Machine Language Programs written in machine language can be executed very fast by the computer. This is mainly because machine instructions are directly understood by the computer and no translation of the program is required. However, writing a program in machine language has several disadvantages, which are discussed below: 1. Machine dependent- Because the internal design of every type of computer is different from every other type of computer, the machine language also differs from computer to computer. Hence, suppose after becoming proficient in the machine language of a particular computer, a company decides to change to another computer, its programmers will have to learn a new machine language and would have to rewrite all the existing programs. 2. Difficult to program. Although machine language programs are directly and efficiently executed by the computer, it is difficult to program in machine language. It is necessary for the programmer either to memorize the dozens of operation code numbers for the commands in the machine's instruction set or to constantly refer to a reference card. A programmer is also forced to keep track of the storage locations of data and instructions. Moreover, a machine language programmer must be an expert who knows about the hardware structure of the computer. 3. Error prone. For writing programs in machine language, since a programmer has to remember the opcodes and must also keep track of the storage locations of data and instructions, it becomes very difficult for him/her to concentrate fully on the logic of the problem. This frequently results in making errors while programming in machine language. 4. Difficult to modify. It is difficult to correct or modify machine language programs. Checking machine instructions to locate errors is very difficult and time consuming. Similarly, modifying a machine language program at a later date is so
difficult that many programmers would prefer to code the new logic afresh instead of incorporating the necessary modifications in the old program. In short, writing a program in machine language is so difficult and time consuming that it is rarely used today. ASSEMBLY LANGUAGE We saw above that programming in machine language is difficult and error-prone mainly because of the following reasons: 1. A programmer needs to write numeric codes for the instructions in the computer's instruction set. 2. A programmer needs to write the storage locations of data and instructions in numeric form. 3. A programmer needs to keep track of the storage locations of data and instructions while writing a machine language program. Introduction of assembly language in 1952 was the first step towards making programming task easier. This was m achieved by overcoming the above listed limitations of machine language programming in assembly language in the following manner: 1. By using alphanumeric mnemonic codes instead of numeric codes for the instructions in the instruction set. For example, using ADD instead of 1110 (binary) or 14 (decimal) for the instruction to add, using SUB instead of 1111 (binary) or 15 (decimal) for the instruction to subtract, and so on. With this feature, the instructions in the instruction set can be much easily remembered and used by the programmers. 2. By allowing storage locations to be represented in the form of alphanumeric addresses instead of numeric addresses. For example, the memory locations 1000, 1001 and 1002 may be represented as FRST, SCND and ANSR respectively in an assembly language program. With this feature, a programmer can much easily remember and use the storage locations of the data and instructions used in an assembly language program. 3. By providing additional instructions, called pseudo-instructions, in the instruction set, which are used for instructing the system how we want the program to be assembled inside the computer's memory. For example, there may be pseudo-instructions for telling the system things like: STAR T STAR T SET SET
PROGRA M DATA
AT 0000
ASIDE ASIDE
AN ADDRESS FOR FRST AN ADDRESS FOR SCN
AT 1000
SET
ASIDE
D AN ADDRESS FOR ANS R
With this feature, a programmer need not keep track of the storage locations of the data and instructions while writing an assembly language program. That is, he/she need not even tell the computer where to place each data item and where to place each instruction of a program. A language, which allows instructions and storage locations to be represented by letters and symbols instead of numbers, is called an assembly language or symbolic language. A program written in an assembly language is called an assembly language program or a symbolic program. Assembler During the discussion of machine language, we saw that a computer can directly execute only machine language programs that use numbers for representing instructions and storage locations. Hence an assembly language program must be converted (translated) into its equivalent machine language program before it can be executed on the computer. This translation is done with the help of a translator program known as an assembler. The assembler of a computer system is a system software, supplied by the computer manufacturer that translates an assembly language program into an equivalent machine language program of the computer. It is so called because in addition to translating an assembly language program into its equivalent machine language program, it also, "assembles" the machine language program in the main memory of the computer and makes it ready for execution. The process of translating an assembly language program into its equivalent machine language program with the use of an assembler is illustrated in Figure 12.2. As shown in the figure, the input to the assembler is the assembly language program (often referred to as a source program) and its output is the machine language program (often referred to as an object program). Since the assembler translates each assembly language instruction into an equivalent machine language instruction, there is a one-to-one correspondence between the assembly language instructions of a source program and the machine language instructions of its equivalent object program. Note that during the process of translation of a source program into its equivalent object program by the assembler, the source program is not being executed. It is only being converted into a form that can be executed by the computer's processor. Assembly language program -► input Assembler -► Machine language program
(Source Program) -► One-to-one correspondence -► (Object Program) Figure 12.2. Illustrating the translation process of an assembler. You must have noticed from the discussion above that in case of an assembly language program, the computer not only has to run the program to get the answer, but it also must first run the assembler (program) to translate the original assembly language program into its equivalent machine language program. This means that the computer has to spend more time in getting the desired answer from an assembly language program as compared to a machine language program. However, assembly language programming saves so much time and effort of the programmer that the extra time and effort spent by the computer is worth it. Assembly language programming and translation of an assembly language program into its equivalent machine language program can be best illustrated with the help of an example. For this, let us assume that the computer uses the mnemonics given in Figure 12.3 for the operation codes mentioned against each. For simplicity, here we have considered only five operation codes that will be used in writing our example program. Like this there can be more than hundred operation codes available in the instruction set of a particular computer. Mnemonic
Opcode
Meaning
HLT 00 Halt, used at the end of program to stop CLA 10 Clear and add into A register ADD 14 Add to the contents of A register SUB 15 Subtract from the contents of A register STA 30 Store A register Figure 12.3. A subset of the set of instructions supported by a computer. We will write a simple assembly language program for adding two numbers and storing the result. The program is shown in Figure 12.4. To get an idea of how the assembler will convert this program into an equivalent machine language program, let us follow its instructions one-by-one. Notice that the first five instructions of the program are pseudo-instructions for telling the assembler what to do. They are not part of the main program to add the two numbers. The first instruction of the assembly language program tells the assembler that the instructions for the ma program (to add two numbers) should start at memory location 0000. Based on this directive, the assembler w load the first instruction for the main program (which happens to be CLA FRST in this example) at memo location 0000 and each following instruction will be loaded in the following
address (that is, ADD SCND will loaded at location 0001, STA ANSR at location 0002, and HLT at location 0003). START PROGRA AT 0000 M START DATA AT 1000 SET ASIDE AN ADDRESS FOR FRST SET ASIDE AN ADDRESS FOR SCND SET ASIDE AN ADDRESS FOR ANSR CLA FRST ADD SCND STA ANSR HLT Figure 12.4. A sample assembly language program for adding two numbers and storing the result. The second instruction of the assembly language program tells the assembler that the data of the program should start at memory location 1000. The next three instructions tell the assembler to set aside addresses for data items FRST, SCND and ANSR. Based on these four directives, the assembler sets up a mapping table somewhere in the computer memory, which looks something like the one shown in Figure 12.5. That is, the assembler picks up the first free address in the data area of the program, which is location 1000, and calls it FRST; it picks up the next free address in the data area, which is location 1001, and calls it SCND; and finally it picks up the next free address in the data area, which is location 1002, and calls it ANSR. Symbolic name
Memory location
FRST 1000 SCND 1001 ANSR 1002 Figure 12.5. Mapping table set up by the assembler for the data items of the assembly language program of Figure 12.4. The next instruction of the assembly language program is CLA FRST, which the assembler translates into 10 1000 by translating CLA into 10 with the help of Figure 12.3 and FRST into 1000 with the help of Figure 12.5. Similarly, the assembler will translate the instruction ADD SCND into 14 1001, and the instruction STA ANSR into 30 1002. Finally it translates the next instruction HLT into 00, thus providing the complete machine language program for the given assembly language program. The resulting machine language program is shown in Figure 12.6.
Memory location 0000 0001 0002 0003 1000
Contents Comments Opcode Address 10 1000 Clear and add the number stored at FRST to A register 14 1001 Add the number stored at SCND to the contents of A register 30 1002 Store the contents of A register into ANSR 00 Halt
Reserved for FRST
1001 Reserved for SCND 1002 Reserved for ANSR Figure 12.6. The equivalent machine language program for the assembly language program given in Figure 12.4. Advantages of Assembly Language over Machine Language Assembly languages have the following advantages over machine languages: 1. Easier to understand and use. Due to the use of mnemonics instead of numeric op-codes and symbolic names for data locations instead of numeric addresses, assembly language programs are much easier to understand and use as compared to machine language programs. Assembly language programming also saves a lot of time and effort of the programmer because it is easier to write as compared to machine language programs. 2. Easier to locate and correct errors. Because of the use of mnemonic op-codes and symbolic names for data locations, and also because programmers need not keep track of the storage locations of the data and instructions, fewer errors are made while writing programs in assembly language, and those that are made are easier to find and correct. Additionally, assemblers are designed to automatically detect and indicate errors for use of an invalid mnemonic op-code or a name that has never been defined. For example, suppose an assembly language program instruction reads ADD AREA, and we forget to define AREA in the program, the assembler will look through its table to find AREA and not finding it will indicate the error.
3. Easier to modify. Because they are easier to understand, it is easier to locate, correct, and modify instructions of an assembly language program than a machine language program. Moreover, insertion or removal of certain instructions from the program does not require change in the address part of the instructions following that part of the program. This is required in case of machine language. 4. No worry about addresses. One of the greatest advantage of assembly language is that programmers need not keep track of the storage locations of the data and instructions while writing an assembly language program. The importance of this advantage can be best illustrated with an example. Suppose we have written a long machine language program involving many instructions and several references to itself within the program, such as looping, branching, and so on. At the very end we may suddenly discover that we have left out an instruction in the middle. If we insert that instruction, we will have to go through the entire program to check and modify (if necessary) all references to other instructions. This is certainly a tedious job. But if we write the same program in assembly language, we merely add the extra instruction, and the assembler will take care of automatically modifying the references suitably. 5. Easily relocatable. The availability of pseudo-instructions for instructing the system how we want the program to be assembled inside the computer's memory makes assembly language programs easily relocatable because their location can be easily changed merely by suitably changing the pseudo-instructions. This is not easily possible with machine language programming. This feature can be best illustrated with the help of an example. Suppose that an assembly language program starts at address 1000 and we suddenly find that we have another program to be used with this program and this program also starts at location 1000. Obviously, one of the two programs will have to be rewritten to be moved to somewhere else. In machine language, this can be a complicated job. But in case of assembly language, we merely have to suitably change the pseudo-instructions for assembling the program for one of the two programs. For example, we change the following pseudo-instructions for one of the programs: to the following pseudo-instructions: START PROGRAM AT 1000 START DATA AT 2000 START PROGRAM AT 3000 START DATA AT 4000 and run the assembly language program once again through the assembler. The equivalent machine language program will this time start at memory location 3000 instead of 1000, and there will be no conflict with the other program. In other words, assembly language programs can be easily moved from one section of the memory to another.
6. Efficiency of machine language. In addition to the above mentioned advantages, an assembly language program also enjoys the efficiency of its corresponding machine language program because there is one-to-one correspondence between the instructions of an assembly language program and its corresponding machine language program. In other words, an assembly language program will be just as long as the resulting machine language program. So leaving out the translation time required by the assembler, the actual execution time for an assembly language program and its equivalent machine language program (written independently) will be the same. Limitations of Assembly Language The following limitations of machine language are not solved by using assembly language: 1. Machine dependent. Because each instruction of an assembly language program is translated into exactly one machine language instruction, assembly language programs are machine dependent. That is, assembly languages differ from computer to computer and an assembly language program can be executed only on the computer in whose assembly language it has been written. Thus, a decision to change to another computer will require learning a new language and the conversion of all existing programs into the assembly language of the new computer. 2. Knowledge of hardware required. Since assembly languages are machine dependent, an assembly language programmer must have a good knowledge of the characteristics and the logical structure of his/her computer in order to write good assembly language programs. 3. Machine level coding. In case of an assembly language, instructions are still written at the machine-code level. That is, one assembly language instruction is substituted for one machine language instruction. Thus writing assembly language programs is still time-consuming and not very easy. Typical Uses of Assembly Language Due to the above mentioned limitations of assembly languages, programmers rarely write programs of any significant size in assembly language today. Today assembly language programming is done only when a program's execution efficiency is very important. Assembly language programming helps in producing compact, fast, and efficient code because it gives the programmer total control of the computer's CPU, memory and registers. Hence, assembly language programming is mainly used today to fine-tune important parts of programs written in a higher-level language. High-level language programs that are candidates for this type of fine-tuning are those whose execution consumes a substantial part of the total computing resources at their installations, and hence
offer real payoffs for any improvement in execution time. In such circumstances, a clever programmer who is familiar with the compiler-generated code for the highlevel language program, and also with the purpose of the program at hand, is very likely to be able to improve it significantly by rewriting in assembly language some small but critical section of it. It would not be very unusual for the programmer to be able to improve the program’s performance by an order of magnitude by rewriting a few dozen, or a few hundred, assembly language instructions. Assembly Languages with Macro Instructions In general, assembly languages are termed one-for-one in nature, that is, each assembly language instruction will result in one machine language instruction. However, quite often a certain set of machine language or assembly language instructions have to be used over and over. For example, three instructions, one after the other, might be needed to print out a value on a particular computer. These three instructions, always in the same order, might be used over and over in the same program. Instead of forcing the programmer to write out the set of three instructions every time he/she wants to print a value, we might as well design the assembler (program) in such a way so as to take care of these instructions. Every time the programmer gave the PRINT instruction, for example, the assembler would translate it into three machine language instructions instead of one, thus supplying the complete set of instructions required for printing. Any instruction, such as PRINT, which gets translated into several machine language instructions, is called a macro instruction. There might be many such macro instructions permitted by a particular assembler. Thus, to speed up the coding process, assemblers were developed that could produce a variable amount of machine language instructions for each macro instruction of the assembly language program. The use of macro instructions adds much work to the computer because the translation process becomes more than just changing each word into a number. The assembler must be able to supply the missing steps as well, but it means a tremendous saving of work for the programmer. The programmer gets relieved of the task of writing an instruction for every machine operation performed. Thus the use of macro instructions reduces the length of the programs, cuts down on errors, and simplifies programming. The macro instruction capability was developed very early in the evolution of computer languages. In fact, it is this concept of multiple machine instructions from one macro instruction around which today's machine-independent high-level languages are designed.
HIGH-LEVEL LANGUAGE In the discussion above we saw that both machine and assembly languages have the following limitations: 1. They are machine dependent. A machine language program or an assembly language program cannot be executed on any computer other than the one for which it is written. 2. They require the programmers to have a good knowledge of the internal structure of the computer being used. Hence only people with sound hardware knowledge could become good machine language or assembly language programmers. 3. It is difficult, error prone and time consuming to write programs in machine language or assembly language because they deal with machine-level coding requiring one instruction to be written for each machine-level operation. Due to these limitations, machine and assembly languages are often referred to as low-level languages. High-level programming languages were designed to overcome the above listed limitations of low-level programming languages. That is, high-level languages are characterized by the following features: 1. They are machine independent. That is, a program written in a high-level language can be easily ported and executed on any computer that has the translator software for the high-level language. 2. They do not require the programmers to know anything about the internal structure of the computer on which the high-level language«*programs will be executed. In fact, since high-level languages are machine independent, a programmer writing a program in a high-level language may not even know on which computer will his/her program be executed. This allows the programmers to mainly concentrate on the logic of the problem rather than be concerned with the details of the internal structure of the computer. 3. They do not deal with the machine-level coding. Rather they deal with highlevel coding enabling the programmers to write instructions using English words and familiar mathematical symbols and expressions. Each statement of a highlevel language is normally a macro instruction that gets translated into several machine language instructions. This is one-to-many translation and not one-to-one as in the case of assembly language. For example, let us consider the same problem of adding two numbers (FRST and SCND) and storing the sum in ANSR. We have already seen that three low-level (machine/assembly) instructions are required for performing this job. However, if we use a high-level language, say FORTRAN for instance, to instruct the computer to do this job, only one instruction need be written: ANSR = FRST + SCND
This instruction is obviously very easy to understand and write because it resembles the familiar algebraic notation for adding two numbers: a = b + c. The advent of high-level languages has enabled the use of computers to solve problems even by non-expert users. This has allowed many users, without any background in computer science and engineering, to become computer programmers. This in turn has resulted in the creation of a large number of computer applications in diverse areas, leading to the use of computers today in every walk of life. Compiler We saw that a computer can directly execute only machine language programs. Hence a high-level language program must be converted (translated) into its equivalent machine language program before it can be executed on the computer. This translation is done with the help of a translator program known as a compiler. Thus, a compiler is a translator program (much more sophisticated than an assembler) that translates a high-level language program into its equivalent machine language program. A compiler is so called because it compiles a set of machine language instructions for every program instruction of a high-level language. The process of translating a high-level language program into its equivalent machine language program with the use of a compiler is illustrated in Figure 12.7. As shown in the figure, the input to the compiler is the high-level language program (often referred to as a source program) and its output is the machine language program (often referred to as an object program). Since high-level language instructions are macro instructions, the compiler translates each highlevel language instruction into a set of machine language instructions rather than a single machine language instruction. Hence there is one-to-many correspondence between the high-level language instructions of a source program and the machine instructions of its equivalent object program. Note that during the process of translation of a source program into its equivalent object program by the compiler, the source program is not being executed. It is only being converted into a form that can be executed by the computer's processor. High-level language program -►input -►complier-►output-►Machine language program (Source Program) -► One-to-many correspondence-► (Object Program) A compiler can translate only those source programs, which have been written in the language for which the compiler is meant. For example, a FORTRAN compiler is only capable of translating source programs, which have been written
in FORTRAN. Therefore, each computer requires a separate compiler for each high-level language that it supports. That is, to execute both FORTRAN and COBOL programs on a computer, this computer must have a FORTRAN compiler and a COBOL compiler. This is illustrated in Figure 12.8. Program PI in high-level language LI -►Compiler for language LI -► Machine code for PI Program P2 in high-level language L2 -►Compiler for language L2-► Machine code for P2 A computer supporting languages LI and L2 Figure 12.8. Illustrating the requirement of a separate compiler for each high-level language supported by a computer. Also note that an object program for one computer will not be the same as the object program for another computer. Hence it is necessary that each computer must have its own "personal" compiler for a particular language, say LI. Figure 12.9 illustrates how machine-independence is achieved by using different compilers to translate the same high-level language program to machine language programs of different computers. Compilers are large programs, which reside permanently on secondary storage. When a source program is to be translated, the compiler and the source program are copied from secondary storage into the main memory of the computer. The compiler, being a program, is then executed with the source program as its input data. It generates the equivalent object program as its output, which is normally saved in a file on secondary storage. Whenever there is a need to execute the program, the object program is copied from secondary storage into the main memory of the computer and executed. Note that there is no need to repeat the compilation process every time you wish to execute the program. This is because the object program stored on secondary storage is already in machine language. You simply have to load the object program from the secondary storage into the main memory of the computer, and execute it directly. Also note that compilation is necessary whenever we need to modify the program. That is, to incorporate changes in the program, you must load the original source program from secondary storage into the main memory of the computer, carry out necessary changes in the source program, recompile the modified source program, and create and store an updated object program for execution. In addition to translating high-level language instructions into machine language instructions, compilers also automatically detect and indicate certain types of errors in source programs. These errors are referred to as syntax errors and are typically of the following types: 1. Illegal characters 2. Illegal combination of characters
3. 4.
Improper sequencing of instructions in a program Use of undefined variable names
A source program containing one or more errors detected by the compiler will not be compiled into an object program. In this case, the compiler will generate a list of coded error messages indicating the type of errors committed. This error list is an invaluable aid to the programmer in correcting the program errors. As shown in Figure 12.10, the programmer uses this error list to re-edit the source program for removing the errors and creates a modified source program that is recompiled. As shown in the figure, the process of editing the source program to make necessary corrections and recompiling the modified source program is repeated over and over again until the source program is free of all syntax errors. Also notice from the figure that the compiler generates the object program only when there are no syntax errors in the source program. A compiler, however, cannot detect logic errors. It can only detect grammatical (syntax errors) in the source program. It cannot know ones intentions. For example, if one has wrongly entered -25 as the age of a person, when he/she actually intended +25, the compiler cannot detect this. Programs containing such errors will be successfully compiled and the object code will be obtained without any error message. However, such programs, when executed, will not produce correct results. So logic errors are detected only after the program is executed and the result produced does not tally with the desired result. START Edit source program Source program Compile source program Syntax errors detected? No Generate object program Object program Stop If yes Generate list of coded error messages Edit source program Figure 12.10. Illustrating recompilation of a source program until it is free of all syntax errors. Linker
A software often consists of several thousands, even several millions, of lines of program code. For software of this size, it is impractical to store all the lines of program code in a single source program file due to the following reasons: 1. The large size of the file would make it very difficult, if not impossible, to work with. For example, it might not be possible to load the file for compilation on a computer with limited main memory capacity. Again, while editing the file, it could be very tedious and time-consuming to locate a certain line of code. 2. It would make it difficult to deploy multiple programmers to work concurrently towards the development of the software for completing it within a specified time limit. 3. Any change in the source program, no matter how small, would require the entire source program to be recompiled. Recompilation of large source programs is often a time-consuming process. To take care of these problems, a modular approach is generally adapted to develop reasonably sized software. In this approach, the software is divided into functional modules and separate source programs are written for each module of the software. Often there is no need to even write source programs for some of the modules because there might be programs available in a program library, which offer the same functionality. These library programs are maintained in their object code form. When modular approach is used for developing a software, the software consists of multiple source program files. Each source program file can be modified and compiled independent of other source program files to create a corresponding object program file. In this case, a program called a linker is used to properly combine all the object program files (modules) of the software, and to convert them into the final executable program, which is sometimes called a load module. That is, a linker takes object program files (modules) and fits them together to assemble them into the program's final executable form. The process of compiling the multiple source program files of a modularly developed software and linking them to create the executable program (load module) is illustrated in Figure 12.11. Note that since converting the source programs into an executable program is a two-step process, when the software is modified, there is no need to recompile all the source programs. Rather, only the changed source programs need to be recompiled to produce new object programs for them. The linker can then be executed to create a new executable program by reassembling the newly created and the old object programs and library modules. Linkers are supplied as a part of the development environment (which consists of a set of program development tools) of a computer system that includes a compiler. One can also purchase third-party linkers that can sometimes improve the performance of a program over the standard linkers provided as a part of the system's development environment.
Interpreter An interpreter is another type of translator used for translating programs written in high-level languages. It takes one statement of a high-level language program, translates it into machine language instructions, and then immediately executes the resulting machine language instructions. That is, in case of an interpreter, the translation and execution processes alternate for each statement encountered in the high-level language program. This differs from a compiler, which merely translates the entire source program into an object program and is not involved in its execution. As shown in Figure 12.12, the input to an interpreter is the source program, but unlike a compiler, its output is the result of program execution instead of an object program. After compilation of the source program, the resulting object program is permanently saved for future use and is used every time the program is to be executed. So repeated compilation (translation of the source code) is not necessary for repeated execution of a program. However in case of an interpreter, since no object program is saved for future use, repeated interpretation (translation plus execution) of a program is necessary for its repeated execution. Note that since an interpreter translates and executes a high-level language program statement-bystatement, not only the entire program must be reinterpreted for its repeated execution, but even a program statement must be reinterpreted (translated and executed) every time it is encountered during program execution. For example, during the execution of a program, each instruction in a loop will have to be reinterpreted every time the loop is executed. As compared to compilers, interpreters are easier to write because they are less complex programs than compilers. They also require less memory space for execution than compilers. The main advantage of interpreters over compilers is that a syntax error in a program statement is detected and brought to the attention of the programmer as soon as the program statement is interpreted. This allows the programmer to make corrections during interactive program development. Therefore, interpreters make it easier and faster to correct programs. The main disadvantage of interpreters over compilers is that they are slower than compilers when running a finished program. This is because each statement is translated every time it is executed from the source program. In case of a compiler, each statement is translated only once and saved in the object program. The saved object program can be executed many times as and when needed, and no translation of any statement is required during the execution of the program.
Because the interpreter does not produce an object program, it must perform the translation process each time a program is executed. To combine the advantages of both interpreters and compilers, sometimes the prograrn development environment of a computer system provides both a compiler and an interpreter for a high-level language. In these cases, the interpreter is used to develop and debug programs. Then, after a bug-free state is reached, the programs are compiled to increase their execution speed. The object program produced by the compiler is subsequently used for routine processing. Assemblers, compilers, and interpreters are also referred to as language processors since they are used for processing a particular language instructions.
Advantages and Limitations of High-level Languages Advantages of High-level Languages High-level languages enioy the following advantages over assembly and machine languages: 1. Machine independence. A program written in a high-level language can be executed on many different types of computers with very little or practically no effort of porting it on different computers. This means that a company changing computers, even to one from a different manufacturer, will not be required to rewrite all the programs that it is currently using. This also means that a commercial software will have a larger market because it need not be restricted to one type of computer. Thus, the time and effort spent on software development gets better rewarded with high-level language programming. 2. Easier to learn and use. High-level languages are easier to learn because they are very similar to the natural languages used by us in our day-to-day life. They are also easier to use because a programmer need not know anything about the internal structure of the computer for programming in a high-level language. 3. Fewer errors. While programming in a high-level language, a programmer need not worry about how and where to store the instructions and data of the program, and need not write machine-level instructions for the steps to be carried out by the computer. This allows the programmer to concentrate more on the logic of the program under development. All these factors lead to fewer programming errors during program development. Furthermore, compilers and interpreters are designed to automatically detect and indicate syntax errors.
Hence syntax errors, if any, in the program can be easily located and corrected by the programmer. 4. Lower program preparation cost. Writing programs in high-level languages requires less time and effort, which ultimately leads to lower program preparation cost. Generally, the cost of all phases of program preparation (coding, debugging, testing, etc.) is lower with a high-level language than with an assembly language or with a machine language. Moreover, the returns on the investment made in developing a software are more when the software is developed using a high-level language because this makes the same software usable on many different types of computers. 5. Better documentation. The statements of a program written in a high-level language are very similar to the natural language statements used by us in our dayto-day life. Hence they can be easily understood by a programmer familiar with the problem domain. As a result, very few or practically no separate comment statements are required in programs written in high-level languages. Due to this reason, high-level languages are also sometimes referred to as self-documenting languages. 6. Easier to maintain. Programs written in high-level languages are easier to maintain than assembly language or machine language programs. This is mainly because they are easier to understand and hence it is easier to locate, correct, and modify instructions as and when desired. Insertion or removal of certain instructions from a program is also possible without any complication. Thus, major changes can be incorporated with very little effort. Limitations of High-level Languages The two main limitations of high-level languages are as follows: 1. Lower efficiency. Generally, a program written in a high-level language has lower efficiency than one written in an assembly language or a machine language to do the same job. That is, programs written in high-level languages take more time to execute and require more main memory space. Hence, when a program's efficiency is very important, its performance critical parts are written in assembly language and non-critical parts are written in a high-level language. 2. Less flexibility. Generally, high-level languages are less flexible than assembly languages because they do not normally have instructions or mechanism to control the computer's CPU, memory and registers. An assembly language provides the programmers access to all the special features of the machine they are using. Certain types of operations, which are easily programmed using the machine's
assembly language, are impractical to attempt using a high-level language. This lack of flexibility means that some tasks cannot be done in a high-level language, or can be done only with great difficulty. In most cases, the advantages of high-level languages far outweigh the disadvantages. Most computer installations use a high-level language for most programs and use an assembly language for doing special tasks that cannot be done easily otherwise. OBJECT-ORIENTED PROGRAMMING LANGUAGES What is Object-Oriented Programming (OOP)? Soon after the advent of first few programming languages, it was realized by the people working in the computer industry that programming is a difficult and time consuming task. Hence researchers have been continuously working towards developing better and better programming languages to make programming simpler, easier, and faster. The goal of the designers of every new programming language was to introduce some aspects or concepts in their programming language that will help in achieving the above mentioned objectives. ObjectOriented Programming (often referred to as OOP, pronounced "oop") is one such concept used in programming languages. The concept of OOP was first introduced in 1967 by the developers of a programming language named Simula67. This concept started gaining popularity in 1980s with the development of another programming language named Smalltalk, which made extensive use of this concept. Today this concept has been made part of almost all major programming languages and is being used quite widely in software development. The basic idea behind OOP is that programming languages are used for simulating real-world problems c computers. Because much of the real world is made up of objects, a simulation of such a world must include simulated objects. Hence the essence of OOP is solving problems by identifying the real-world objects of the problem and the processing required of those objects, and then creating simulations of those objects, the processes, and the required communications between the objects. Fundamental Concepts of OOP The most essential part of support for OOP is the object mechanism. That is, an OOP language must ha^ mechanisms to define, create, store and manipulate objects, and some mechanism to allow different objects communicate with each other. The basic concepts used in OOP languages to support the object mechanism a: described below. Object
An object is the primitive element of a program written in an OOP language. Each object consists of a set < procedures (known as methods) and some data. It stores the data in variables and responds to messages received from other objects by executing the procedures (methods). A method is a collection of operations that specifics the reaction of the object when it receives a message that corresponds to that method. A program written in a pure OOP language is a collection of objects that communicate with each other. Virtually everything, from items as simple as the integer constant 8 to a complex file-handling system, are objects. As objects, they are treated uniformly. They all have local memory, inherent processing ability, the capability to communicate with other objects, and the ability to inherit characteristics from ancestors. Each object can be viewed as an abstract data type because it is an autonomous entity with its private data and methods. Note that an object contains both functions (procedures/methods) and data that describe the characteristics of the object. In other words, an object encapsulates both data and functions. Method A method of an object defines the set of operations that the object will execute when a message corresponding to the method is received by the object. In a sense, methods are like function definitions, complete with parameters and the capability of returning values. The binding of a message to a method is dynamic. The entire collection of methods of an object is called the message protocol, or message interface, of the object. Figure 12.13 illustrates the object model used in OOP.
Message The mechanism to support communication between objects is through messages. That is, all computing in a program written in an OOP language is done by sending a message to an object to invoke one of its methods. A reply to a message is an object, which returns the requested information, or simply notifies the sender that the requested processing has been completed. There is a fundamental difference between a message sent to an object and a subprogram call. In the former case, the object executes the appropriate method
with its own local data, whereas in the latter case the data to be processed is usually sent along with the subprogram call to the subprogram code unit. Class A class is a description of one or more similar objects. The class concept is similar to "type" in the conventional procedure-oriented languages. For example, if "integer" is a type (class), then "8" is an instance (object) of type (class) "integer". Similarly, "Person-1" and "Person-2" can be two instances (objects) of a class "Person". Multiple instances of a class can be created which are known as the objects of the class. Each object has its own local data and represents a different instance of its class. Since all the objects of a class share the same methods, the only difference between two objects of the same class is the state of their local variables. Also note that there can be two categories of object variables - class variables and instance variables. Class variables are variables stored in the class whose values are shared by all instances (objects) of the class. Instance variables are variables for which local storage is available in the instances (objects). Object-oriented programming languages also support class hierarchies. Subclasses of a given class are refinements of it, inheriting the functionality and local variables of the parent-class, or super-class. Sub-classes can add new local variables and methods and can modify or hide inherited methods. Inheritance In OOP, inheritance is a mechanism for sharing code and behavior. It allows a programmer to reuse the behavior of a class in the definition of new classes. We saw above that a program written in an object-oriented language may have a large hierarchy of classes. In fact, good programming in an object-oriented language depends, to a large extent, in creating sub-classes of existing classes because this leads to potential reuse of existing code. In a hierarchy of classes, a sub-class inherits all of the instance variables, instance methods, and class methods of its super-class/parent-class. The sub-class can also have its own instance variables, which must have different names than the variable names in its ancestor classes. Finally, the sub-class can define both new methods and methods that already exist in an ancestor class. When a sub-class has a method whose name or protocol is the same as an ancestor class, the sub-class method hides that of the ancestor class. Access to such a hidden method is normally provided by prefixing the message with the pseudo-variable "super". This causes the method search to begin in the super-class rather than locally.
Inheritance need not get limited to one parent-class. The child class can inherit from multiple parent classes. This is called multiple inheritance. Figure 12.14 describes persons by an inheritance hierarchy of classes (representing behaviors). The person class has male and female as its sub-classes. The male class has person as its super-class and boy and man as its sub-classes. The instances Ram, Shyam, Mohan, Seeta, Geeta each have a unique base class. Figure 12.15 illustrates the concepts of object, class and inheritance with the help of an example. In the exampl object 'a' is an instance of class 'A' and object 'b' is an instance of class 'B'. Both A and B define behavior t operations shared by their instances, and have instance variables that cause a private copy to be created for each instance for the class or sub-class. The instance 'a' of class 'A' has a copy of A's instance variables and a pointer to its base class. The instance 'b' of class 'B' has a copy of the instance variables of both class 'B' and its superclass 'A' and a pointer to the base class of 'B'. The representation of class 'B' has a pointer to its super-class 'A', while class 'A' has no super-class pointer, since it has no superclass. When object 'b' receives a message to execute a method, it looks first in the methods of class 'B'. If found, the method is executed using the instance variables of 'b' as data. Otherwise, the pointer to its super-class is followed. If it finds the method in class 'A', it executes it on the data of 'b'. Otherwise, it searches A's superclass if there is one. If 'A' has no super-class and the method has not been found, an error is reported. Procedure-oriented versus Object-oriented Programming Conventional procedure-oriented programming languages use procedural abstractions that encapsulate sequences of operations into procedures (also known as functions, subroutines, or subprograms). On the other hand, object-oriented programming languages encapsulate data as well as sequences of operations into abstract entities called objects, providing a stronger encapsulation mechanism than procedures. Hence object-oriented programming provides a more powerful modeling tool than procedure-oriented programming. Notice that both procedures and objects can be considered to be server modules that may be called by clients to get certain actions performed. However, the basic difference between the two is in their roles. The role of procedures is to transform input data specified by parameters into values, while the role of objects is to serve as a repository of data (the current system state) and to respond (perform the requested action) in a manner determined by the current system state. For
example, the response of a bank to a withdrawal request depends on the value of the current balance. The procedure-oriented programming paradigm has strong organizing principles for managing actions and algorithms, but has weak organizing principles for managing shared data. Object-oriented programming paradigm has strong organizing principles both for managing actions and algorithms and for managing shared data because it provides the capability to restrict applicable operations to those associated with a specific object or class. Inheritance provides further strength to this capability by allowing classes to be classified by factoring out properties common to several sub-classes into a super-class. Advantages of OOP Object-oriented programming paradigm typically has the advantages mentioned below over conventional procedure-oriented programming paradigm. Natural Modeling OOP resembles the real worlds more closely than the conventional procedural programming. Hence it serves as more natural modeling tool for programming real world problems. Object-oriented design (often referred to a OOD) involves the identification and implementation of different classes of objects and their behaviors. The objects of the system closely correspond to the real world objects, as is observed and understood by the brain The effort required in the conventional procedural languages, for the translation or transformation of the fact perceived by the brain into some other form suitable to the system, gets greatly reduced with OOP. Hence designing and implementing a system is much easier in OOP. Modular Design In OOP, objects are autonomous entities in the sense that the behavior of each object is defined and implemente in the object itself and is independent of the implementation of other objects. This independence of each object naturally provides the facility of modular system design. Object-oriented design (OOD), is an approach to software design in which the modular decomposition of a system is based on the classes of objects the system manipulates, not on the function the system performs. Modular design eases the development and maintenance of software systems. Abstraction
Abstraction is the principle that states that programs should not make assumptions about implementations and internal representations. In OOP, each object can be viewed as a shielded and encapsulated entity because its data is private to itself (hidden from other objects) and its interaction with other objects is only through its message interface. Any object can be requested for any required information or service by sending the suitable message t it. The requester gets what it asked for without the need to bother about how the object provides the information or service. The 'how' part, that is, the implementation details are internal to the object, the supplier. Thus OOP provides a high degree of abstraction, which gives an opportunity to the programmers to concentrate more o problem specification rather than be taken away by the implementation details. Dynamic Binding In conventional programming languages, the operators, functions, etc. are bound to their respective operations at the time of compilation. This is called static binding. However, in OOP, the binding of operator to an operation takes place at run time. This is called dynamic binding. To have an understanding of this concept, let us consider the following situation. A class named 'shape' has been defined along with a collection of sub-classes for different kinds of shapes, such as circles, triangles, rectangles, etc. If objects of these sub-classes need to be displayed, the display method must be defined within the class of the particular kind of shape, for each kind o shape will require different display code. Suppose that a linked list of objects of various shapes is constructed and code is written to display each object on the list. Then the message to 'display shape' is the same for each object on the list, but it must be dynamically bound to the 'display_shape' method in the proper object. Naturally the same message 'display_shape' sent to different objects has got different responses, which is term& polymorphism. Languages that provide this sort of dynamic binding are called polymorphic languages. Dynamic binding allows abstract data types to be truly generic because it enables a single operator to have more than one operation. Code Reusability In OOP, inheritance provides an effective method of code reuse because the common operations defined and implemented in a class can be accessed by all the classes and instances down the whole class hierarchy. That is, as and when new classes inherit from existing classes, the code developed in super-classes gets reused. This feature leads to tremendous saving of programming effort. Ease of Maintenance
The modular design and code reusability features also help in making software systems easy to maintain. This is because by making a change in a super-class, in effect all its sub-classes are changed at once, without the tedium of identifying and implementing the same changes at a number of places. Moreover, if changes are made in one sub-class to get distinctions, the code in the super-class and the other sub-classes remains safe. Evolution of OOP Languages The first object-oriented programming language, named Simula 67, was released in 1967. Its language primitives included objects, classes, and inheritance. It was used extensively for simulation and other applications, primarily in Europe. However, the programming language community did not pay too much attention to OOP until the Software Concepts Group at Xerox PARC (Palo Alto Research Center) released a language known as Smalltalk in the 1970s. Smalltalk 80, a version of Smalltalk released in 1980, caught the public imagination because of its implementation on personal computers and its interactive graphical interface that permitted browsing and the display of objects in multiple windows. Smalltalk was actually the basis for both the Apple Macintosh graphical user interface and Windows. Starting in the mid-1980s, OOP became a popular term, and object-oriented dialects of existing programming languages began to appear, like Object-Pascal, Objective C, and C++. C++ has proven to be a popular object-oriented programming language because of its connection with C and its support by AT&T Bell Labs. The OOP philosophy has now got extended to almost all popular programming languages including Visual Basic and Java. SOME HIGH-LEVEL LANGUAGES A large number of high-level languages have been developed since the first ones in the early 1950s. By now, more than 1000 high-level languages have been developed, but many of them are no longer in use. Only a handful have been truly significant, and even fewer have been widely used. In this section, we will briefly describe some popular high-level languages. Our primary objective here is to provide some insight into these languages rather than to provide detailed knowledge required to write programs in these languages. Why So Many High-level Languages? Since high-level languages are machine independent, you must be wondering why so many high-level languages were developed, and why many of them are in use today, instead of developing and improving just one good high-level language. For
an answer to this question, recall that one of the objectives of high-level languages is that they should be easier to learn and use by the intended users. Computer programs are written to solve a very diversified class of problems, such as, systems-related problems, scientific and engineering problems, business and commercial problems, graphics problems, etc. The language statements and notations used in a particular problem domain may be quite different from those used in other problem domains. If we try to incorporate all types of statements and notations for solving all classes of problems in a single high-level language, the features of that high-level language will become too vast and difficult to master. Such a language will be very difficult to learn and use. Moreover, designing an efficient compiler for such a language will also be a very tedious task. Hence it is more logical to have different high-level languages for different problem domains. The statements and notations used in a particular high-level language are especially picked up to be useful for solving particular types of problems. For example, a""scientist using a science-oriented language can use scientific formulae, and a businessman with a business-oriented language can use business terms. This makes a high-level language, for a particular problem domain, easier to learn and use by the users dealing with problems in that problem domain. Hence today many different high-level languages exist and are in use because each was designed for different specific purposes. We will see in the discussion below that different high-level languages have different purposes and areas of applicability. FORTRAN FORTRAN stands for FORmula TRANslation. It is one of the oldest high-level language. It was designed to solve scientific and engineering problems and is currently the most popular language amongst scientists and engineers. It was originally developed by John Backus and his team at IBM (International Business Machine Corporation) for its 704 computer in 1957. Since then it has undergone several revisions and has been evolving into a wider and more useful language with time. The original version of FORTRAN that was developed by John Backus and his team was soon followed by a version that was popular as FORTRAN II. The next popular advanced version was FORTRAN IV. In order to make FORTRAN language machine independent (that is, to allow a program that was written for one computer system to be run on another computer system), FORTRAN IV was standardized by American National Standards Institute (ANSI) in the year 1966. With this, FORTRAN became the first standardized language. In 1977, an updated version of FORTRAN IV, called FORTRAN 77, was released and standardized by ANSI. This version was oriented towards structured
programming approach and had several new features for this. Some of the important ones are: 1. Character string handling. 2. Better constructs for expressing program loops. 3. An IF statement with an optional ELSE clause. FORTRAN 90 is the latest version of FORTRAN standardized by ANSI. It again has many new features as compared to FORTRAN 77. Some of the important ones are: 1. A collection of built-in functions for array operations. These include DOTPRODUCT, MATMUL TRANSPOSE, MAXVAL, MINVAL, PRODUCT, and SUM, whose meanings are obvious from their names to users from the scientific and engineering community. 2. Facility to dynamically allocate and deallocate arrays. This is a radical departure from earlier FORTRANs, which all supported only static data. 3. Use of pointer variables that facilitate two distinct kinds of uses - (a) they allow indirect addressing (can be used to point to other data items), which is heavily used in assembly language programming, and (b) they provide a method of dynamic storage management (programmers can allocate a piece of dynamic storage and set a pointer to its address). 4. Addition of new control statements to facilitate easier programming. These include CASE statement for multiple options selection, EXIT statement for prematurely departing from a loop, and CYCLE statement for transferring control to the bottom of a loop but not out. As FORTRAN language is oriented towards solving problems of a mathematical nature, it has been designed as an algebra-based programming language. Any formula or mathematical relationships that can be expressed algebraically can easily be expressed as a FORTRAN instruction, for example, A = B + C - D. To illustrate the nature of FORTRAN programs, a simple FORTRAN program to compute and print the sum of 10 numbers is given in Figure 12.16. C FORTRAN PROGRAM TO COMPUTE C THE SUM OF 10 NUMBERS SUM = 0 DO 50 I =1,10 READ (5, 10) N SUM = SUM + N 50 CONTINUE WRITE (6, 20) SUM 10 FORMAT (F6.2) 20 FORMAT (IX, 'THE SUM OF GIVEN NUMBERS = ', F10.2) STOP END
It can be seen from the example of Figure 12.16 that a FORTRAN program consists of a series of statements for input/output, calculation, logic/comparison, etc. The words READ, WRITE, DO, CONTINUE, STOP, END, etc. in the statements mean exactly what one would expect. In the given example, the first two lines that begin with the character C are comment statements. Comment statements are used in programs for the purpose of documentation or explanation designed to assist anyone reading the source program listings. Comments do not form a part of the program logic and are ignored by the computer. In the third statement of the program, the value of SUM is initialized to zero. The next statement starts a DO loop. This loop ends in the line having label 50, which is a CONTINUE statement. Inside the loop, values of N are read and added to SUM one by one. After the computer loops 10 times, reading and accumulating the sum of 10 numbers, the computer goes out of the loop and drops down to the next statement. This is the WRITE statement which prints the message: THE SUM OF GIVEN NUMBERS = followed by the computed value of SUM. The next statement, which is a STOP statement tells the computer to stop the execution of the program. Finally, the END statement tells the computer that there are no more instructions or statements in the program. The data for this program is contained in a separate file and is not shown in the program COBOL COBOL stands for COmmon Business Oriented Language. As its name implies, it was designed specifically for business data processing applications. Today it is the most widely used programming language for business-oriented applications. Business data processing applications deal with storing; retrieving, and processing corporate accounting information, and to automate such functions as inventory control, billing, and payroll. The initial work on the design of COBOL language was started in 1959 under the leadership of retired Navy Commodore and Mathematician Grace Hopper. The language specifications of COBOL were later worked out in 1959-1960 by a committee of the Conference on DAta SYstems Languages (CODASYL) as a joint effort of computer users, manufacturers, and the United States Government. After the language specifications were defined by this committee, the various manufacturers wrote the COBOL compilers for their computers. An ANSI COBOL standard was first published in 1968. Revised versions of this standard were published by ANSI in 1974 and 1985, which were known as COBOL 74 and COBOL 85. As long as these standards are followed, a COBOL program can be
run on any computer system with an ANSI COBOL compiler. Visual COBOL is an object-oriented version of the language. COBOL language continues to evolve even today. COBOL 2002, the new COBOL standard, has the following new features: 1. Free format, which allows statements to be positioned anywhere on the line. The older versions of COBOL had the concept of A and B margins. The A margin started at column 8 and was used to start a new division, section, or paragraph. The B margin started at column 12 and was used to start any sentence. Compiler directives allow the programmer to switch between fixed format (of older versions) and free format (of new version) in a single program. 2. Inline comments, which allows comments to be positioned anywhere on a line after the two consecutive characters. In older versions of COBOL, comment statements had to be on a separate line with an asterisk (*) in column 7. 3. Common exception handling, which specifies predefined exceptions that will be raised when a runtime exception occurs. The older versions of COBOL did not prescribe the resultant action for runtime errors such as, subscripting with a value larger than the size of a table. 4. Several new data types, which allow more flexible programming. Some of these are boolean, national, and pointer. A boolean data item can have a value of either 0 or 1. A national data type allows the use of extended character sets, such as Unicode or Kanji, for foreign language support. A pointer data item can be used to point to other data items. 5. Conditional compilation of certain code segments, which allows programmers to tailor programs for different environments. For example, Figure 12.17 illustrates how the >> define and evaluate compiler directives can be used to tailor a program based on the value of operating system, which is obtained from the operating environment. 6. Automatic data validation, which allows the programmer to specify constraints with the data description so that the value of input data gets automatically validated. Figure 12.18 illustrates how constraints can be specified with a data description. 7. Call statement enhancements, which allow COBOL programs to interoperate with other programming languages. With these enhancements, COBOL can now easily make use of application programming interface written for C and other programming languages. >>define operating_system as parameter >>evaluate operating_system >>when "unix" *>>code specific to Unix >>when "win2000" *>code specific to Windows 2000 >>when other display "The specified operating system is unknown to the program"
>> end_evaluate 8. File sharing and record locking, which allow the programmer to describe a file with one of the following sharing attributes: "all other", "no other", and "read only". These sharing attributes describe the types of access that a user will allow for other users who access the file simultaneously. A sharing attribute specified with the "open" statement overrides the type of sharing allowed while the file remains open. Users can choose between automatic or manual record locking for a shared file. 9. Object-oriented features, which allow object-oriented concepts such as class, object, inheritance, etc. to be used while writing COBOL programs. COBOL was designed to have the appearance and structure of a business report written in English. Thus, a COBOL program is constructed from sentences, paragraphs, sections and divisions. All COBOL programs must have the following four divisions: 1. Identification Division, which specifies the program and its author. 2. Environment Division, which specifies the computer and the peripherals used to compile and execute the program. If the program is to be processed on a different computer, this division will have to be rewritten. 3. Data Division, which specifies the structure and format of the input and output data files as well as all storage locations used to hold intermediate results and other values during processing. 4. Procedure Division, which contains the sequence of operations to be performed by the program. The nature of COBOL programs is illustrated in Figure 12.19 with the help of a simple COBOL program to compute and print the sum of given numbers. It can be easily seen from this example that COBOL programs are verbose but easy to read. COBOL is often referred to as a self-documenting language. Self-documenting languages are those that do not require much explanation in order to be understood by someone reading the program instructions. The self-documenting aspect of COBOL is made possible by its following features: 1. Its English-like commands. 2. Its English-like sentences and paragraph structure. 3. Its very generous maximum symbolic field-name length of 30 characters. With a field-name length of up to 30 characters, the name can clearly identify the field and its purposes. COBOL programmer should ensure that the used fieldnames are meaningful, so that the self-documenting feature of the language is not lost. The main drawback of COBOL is that it is a verbose language. Writing a COBOL program is somewhat tedious and time consuming because of all the extra words
needed in its statements. However, over the years, this drawback has been taken care of to a large extent in the following manner: 1. A rich set of libraries of COBOL business applications modules and programs has been made available today by many vendors. COBOL programmers often make use of these library programs in their applications for faster development. 2. Automatic program generators for COBOL programs have also been developed and made available by some vendors. These software packages are capable of generating 70 to 80% of the code needed in a particular application program from the application's specifications. The programmer has to manually write only 20 to 30% of the code needed to tie the program elements together and to create a customized problem solution for the application. The use of a COBOL program generator can significantly reduce the time and effort needed to code new COBOL programs. BASIC BASIC stands for Beginners All-purpose Symbolic Instruction Code. It was developed in the year 1964 by Professor John Kemeny and Thomas Kurtz at Darmouth College in the United States. Their main objectives for developing a new programming language were as follows: 1. It should be an interactive language that permits direct communication between the user and the computer during the preparation and use of programs. 2. It must be easy for non2science students to learn and use the language. In order to meet these objectives, the designers took the following approach in the design of BASIC language: 1. It was designed to use terminals as the method of computer access. Before this, most programs were entered into computers through either punched cards or paper tape. 2. It was designed as an interpreted language (unlike FORTRAN and COBOL which were compiled languages) so that programmers could create, run, test and debug a program in interactive mode. 3. It was designed with very few statements and grammatical rules so that it could be easily learnt and used by beginner programmers. The language translator used for BASIC is an interpreter instead of a compiler. This allows a programmer to check for syntax errors and immediately correct it as the program is entered. It also makes the language simple and easy to implement and use. Because of its simplicity to implement, learn and use, BASIC was the first highlevel language to be implemented on personal computers when they were introduced. Today, BASIC is available in almost all personal computers and even in some pocket calculators. This fact makes it one of the most widely installed computer languages in the world.
Though simple and easy to learn, BASIC is quite flexible and reasonably powerful. It can be used for both business and scientific applications. Users of BASIC range from school students to scientists and engineers to business managers. It is also used to teach programming basics to beginners. A minimal version of BASIC was standardized by ANSI in 1978. But this standard is so simple that it has been extended in virtually every available BASIC dialect. Today, BASIC has many dialects including BASICA, QBASIC, and Visual Basic. Microsoft's Visual Basic supports object-oriented features and methods. Furthermore, note that although BASIC began as an interpreted language, several new dialects of BASIC have added compilers. Microsoft's Quick BASIC and Visual Basic produce compiled programs. A BASIC program to compute and print the sum of 10 numbers is given in Figure 12.20. It can be observed from this example that a BASIC program is made up of a series of statements. Each statement starts with a statement number and a key word, followed in most cases by some type of action. For example, in the statement, "40 LET S = S + N", 40 is the statement number, LET is the key word, and S = S + N is the action. The first two statements of the program, which are REM statements, are remarks being made for the purpose of explaining or documenting a program step. They have no effect on the program logic. A remark/comment statement must have a line number, the key word REM, and any remark that the programmer wishes to make. In our example, the remark statement was used to name the program. The two LET statements are assignment statements. The READ and PRINT statements read input data and print output result respectively. The FOR and NEXT statements control the loop that is meant for adding the 10 numbers one-by-one. The two DATA statements in the program are used to furnish the input data. The END statement stops the program. 5 REM PROGRAM TO COMPUTE 6 REM THE SUM OF 10 NUMBERS 10 LETS = 0 20 FORI=lTO10 30 READ N 40 LET S = S + N 50 * NEXT I 60 PRINT "THE SUM OF GIVEN NUMBERS : 70 DATA 4, 20, 15, 32, 48 80 DATA 12, 3, 9, 14,44 90 END; Pascal Named after the famous seventeenth-century, French mathematician, Blaise Pascal, this language was first developed in the year 1971 by Professor Nicklaus Wirth of Federal Institute of Technology in Zurich, Switzerland. His main
objective was to develop a language that can allow beginners to learn good problem solving and programming practices. To meet this objective, Pascal was developed based on the concepts associated with structured programming. That is, it was designed in such a way that complete Pascal programs could be written without the use of any GO TO statement. Special control statements for selection and looping structures were provided for this purpose. To make programs structured, Pascal programs are composed of blocks. Each block of statements starts with a Begin statement and ends with an End statement. Due to the above mentioned features, Pascal is widely recognized as a language that encourages programmers to write well-structured, modular programs and instills good programming practices in a programmer. Owing to this reason, it is recognized as an educational language and is used to teach programming to beginners. Pascal was standardized by ANSI in 1983. Since then several dialects of Pascal have been made available on different computer systems. An object-oriented version of Pascal is also available. Pascal is suitable for both scientific and business applications because it has features to manipulate, not only numbers, but also vectors, matrices, strings of characters, sets, records, files and lists. A Pascal program to compute and print the sum of 10 numbers is given in Figure 12.21. The first line of the program contains the name of the program which is SUMNUMS. This is followed by two comment lines. Any comment can be placed within the symbols (* and *). Then, all the variables are declared. The variables SUM and N are declared as real and hence they can be assigned any real number. Similarly, the variable I, that has been declared to be an integer variable can be assigned any integer value. The heart of the program starts with the word BEGIN and ends with the word END. First, the variable SUM is initialized to zero. The next statement starts a DO loop that reads and computes the sum of the 10 numbers. Finally, the statement having WRITELN prints the result of the computation. PROGRAM SUMNUMS (INPUT, OUTPUT);
(* PROGRAM TO COMPUTE THE SUM OF 10 NUMBERS *) (* DECLARATION OF VARIABLES *) VAR SUM, N : REAL; VAR I: INTEGER; (* MATN PROGRAM LOGIC STARTS HERE *) BEGIN SUM := 0; FORI:= 1 TO 10 DO BEGIN READ (N); SUM := SUM + N; END; WR1TELN ('THE SUM OF GIVEN NUMBERS=', SUM);
END; PL/1 PL/1 stands for Programming Language One. It was designed by IBM in 1964. The main objective of the designers was to design a universal programming language that could be used for programming any kind of application. Note that during those days, FORTRAN was used for scientific applications and COBOL was used for business applications. To meet the objective mentioned above, PL/1 was designed to have the following features: 1. Features similar to COBOL for programming business applications. 2. Features similar to FORTRAN for programming scientific applications. 3. Other features such as string manipulation, list processing, pointer data type, concurrent execution of tasks and exception handling, for systems programming and other types of applications. With all these features, PL/1 became one of the most versatile and most powerful of the programming languages. However, it could not become the most commonly used language because with so many features included in it, it also became a sophisticated language to learn and use. It is enormous by any standards and is not easy to learn in its totality. Furthermore, the complexity of the language makes a compiler and support packages for the full language quite large. Because of its size and heritage, the language has so far been available primarily on IBM systems and is not very widely used or accepted. A PL/1 standard was produced by ANSI in the year 1976. A subset of this full standard, known as PL/l-G was also made available for use with personal computers. A PL/1 program to compute and print the sum of 10 numbers is given in Figure 12.22. It can be seen from this example, that the basic element of a PL/1 program is a statement. Each statement is terminated with a semicolon (;) and several statements may be combined to form a procedure. A procedure may represent an entire small program (as in this example) or a block or module of a more complex program. Because of its modular structure, a beginner needs only a small part of the language in order to write programs for a particular type of application. Moreover, modular procedure blocks and other features of the language allow the use of the various concepts of structured programming. SUMNUMS: PROCEDURE OPTIONS (MAIN); /* PROGRAM TO COMPUTE THE SUM OF 10 NUMBERS */ DECLARE (SUM, N) FIXED; DECLARE I FIXED; SUM = 0; DO 1= 1 TO 10;* GET (N); SUM = SUM + N; END;
PUT EDIT ('THE SUM OF GIVEN NUMBERS = ', SUM) (A, F5); END; SOME MORE HIGH-LEVEL LANGUAGES The high-level programming languages that were discussed above are not necessarily the most important or most popular languages. These languages were presented in some detail to give you a better understanding of computer programming and nature of high-level languages in general. There are several other programming languages, which are equally important and popular. Some of these languages are described in this section. C and C++ C language was developed in 1972 at AT&T's Bell laboratories, USA by Dennis Ritchie and Brian Kernighan. Their objective was to develop a language which will have the virtues of high-level programming languages (mainly machine independence) with the efficiency of an assembly language. To meet this objective, C language was designed to have the following features: 1. It was developed as a compiled language (uses compiler as its translator) so that C language programs could be easily ported to other computers equipped with a C compiler. 2. It supports pointers with pointer operations. This feature allows the programmers to directly access the memory addresses where variables are stored and to perform bit-level manipulation of data stored in memory or processor registers. Pascal and PL/1 also supported pointers, but restricted their use. C language has minimal restrictions on the use of pointers allowing greater flexibility to programmers in writing efficient programs. 3. It supports user-defined data types for greater flexibility in programming. 4. It supports modular and structured programming concepts. That is, while writing a C program to solve a problem, the problem may be broken down into smaller tasks, a function may be written (or used) to solve each task, and the C program may become a group of functions that are linked together to produce the problem solution. It also supports a rich library of functions, which can be directly used by programmers for efficient coding of their applications. 5. It is a small and concise language providing only the bare essential features so that a C language program can be translated by a language translator into an efficient machine language code. Due to its features mentioned above, C soon became the language of choice of systems programmers who wanted to write portable systems software and commercial software packages like operating systems, compilers, spreadsheet, word processing, and database management systems. In fact, the first major use of C was to write an operating system known as UNIX. Today C is extensively used
by all types of programmers for programming various different types of applications. Although conciseness and flexible use of pointers make C a very powerful language, they also make C a language which is difficult to learn and comprehend. Beginner programmers often find it difficult to understand and use the flexibility offered by pointers in programming a solution. Moreover, its conciseness feature allows brief programs to be written for carrying out complex computations. Often such programs are difficult to understand and modify, particularly when they make extensive use of pointers. For several years after its development, the only standard for C was a book on C written by its developers Kemighan and Ritchie. An ANSI standard for C was produced in 1989. A new version of C name C++ (pronounced "C plus plus") was developed by Bjarne Stroustrup at Bell Labs in the early 1980s. The evolution o1 C++ continued during the first half of 1980, and the first widely available implementation appeared in 1985. C++ is so called because ++ is an operator in C that increments a variable by 1. Therefore, the C language ii incremented to its next level with C++. C++ contains all the elements of the basic C language but has been expanded to include numerous object-oriented programming features. The most essential part of support for object-oriented programming is the class/object mechanism, which is the central feature of C++. C++ provides a collection of predefined classes, along with the capability of user-defined classes. C++ being a superset of C is an extremely powerful and efficient language. But C++ is also more difficult learn than C because learning C++ means learning everything about C, and then learning object-oriented design and its implementation with C++. Over the years, C++ has become a very popular language. It is said that in a few more years C++ may replace because more and more C programmers are migrating to C++ every year. Another factor in favor of the future popularity of C++ is that it is almost completely downward compatible with C (meaning that C programs can be, for the most part, compiled as C++ programs), and in most implementations it is possible to link C++ code with code. Java Java is a language primarily used for Internet-based applications. Its development started at Sun Microsystems 1991 by a team led by James Gosling. The language was formally announced in May 1995, and its j commercial release was made in early 1996.
The main objective of the developers of Java was to design a language with the following features: 1. The compiled code should be machine-independent (recall that the compiled code of other high-level languages is machine dependent). 2. The developed programs should be simple to implement and use even in consumer electronics products like TV set-top boxes, hand-held devices, telephones, and VCRs. Java was designed to have these features in the following manner: 1. Java uses the concept of just-in-time compilation in which Java programs are typically compiled into machine-independent byte-code, which is interpreted by the Java runtime system at program execution time. 2. Java was developed as a small subset of C++ features to keep it simple and easy to use. Several of the complex features of C++ such as pointer and multiple inheritance are not part of Java. Due to this, Java is considered to be a "clean" object-oriented language. Before the innovation of Java, Internet applications were static in nature because they offered visually static pages of information. The advent of Java and its use in Internet applications transformed the static pages into a more dynamic one that allows users to interact with what appears on the screen. Hence, Java brings animation and interactivity to Internet-based applications. The use of Java is not only limited to Internet-based applications. It can also be used to develop anything from spreadsheets to tutorials to interactive games. It can also be used in embedded systems, as in consumer electronics products such as hand-held devices, telephones, and VCRs. Software developed to allow the user to observe and interact with Java programs has been named HotJava. Ada Ada is a multi-purpose language developed for and used primarily by the U.S Department of Defense (DOD) for military applications. It was developed in 1980 at Honeywell Computer Company by a group of scientists headed by Ichbiah. It is named in the honor of Lord Byron's daughter Ada Augustha Lovelace. Ada used to work with Charles Babbage and is considered to be the first programmer because she wrote the first computer program for the mechanical computer (Analytical Engine) developed by Charles Babbage in 1842. Ada was standardized by ANSI in 1983. The main objective of the developers of Ada was to design a language with the following features:
1. It should be suitable for writing applications that deal with multiple concurrent processes because this feature is very common in military applications. 2. The language should have good error-handling capability. 3. The language should support the use of data abstraction in program design. 4. It should be possible to write generic procedures that can be easily reused. Ada was designed to have these features in the following manner: 2. For error handling, Ada has excellent facilities for exception handling, which allows the programmer to gain control after run-time errors have been detected. 3. For data abstraction, Ada uses the concept of packages, which provide the means for encapsulating specifications for data types, data objects, and procedures. Packages are specified or written into the declarative part of the program, which is typically at the top part of an Ada program. Then various procedural statements in the program can access the packages and use it. 4. For generic procedures, Ada allows functions (or subroutines) to be written with parameters whose data type can be decided when the function is to be used. For example, it is possible to write a sort procedure that uses an unspecified type for the data to be sorted. Such a generic procedure must be instantiated for a specified data type before it can be used. This is done with a statement that causes' the compiler to generate a version of the procedure with the given type. The availability of such generic procedures increases the range of program units that might be reused. With all these features, Ada turned out to be an extremely powerful and flexible language. However, with a very large number of features, it also turned out to be an extremely complicated language. The development of a compiler for the language is an extremely difficult task. It is also very difficult for a programmer to master the language because it is too large and too complex. Due to these reasons, Ada could not become a very popular language in-spite of its rich features. ALGOL ALGOL stands for ALGOrithmic Language. As the name implies, ALGOL was designed for specifying algorithms, primarily of scientific nature, which was the primary computer application area at that time. It was designed by an international group comprising of scientists and mathematicians from Europe and the United States. The first version of the language was released in 1958, and it was known as ALGOL 58. Later on, it was revised in 1960 and this revised version, known as ALGOL 60, was the most widely used version of the language. The latest and the most powerful version of the language is ALGOL 68 that was released in 1968. Scientific applications being the main area of use, ALGOL was designed as an extension of FORTRAN. That is. it generalized many of FORTRAN'S features and added several new constructs and concepts to make the language more flexible and powerful. Some of the main features of ALGOL are as follows:
1. It allows variable names to be of any length, as opposed to then FORTRAN'S restriction to six characters. 2. It allows arrays to have any number of dimensions, as opposed to then FORTRAN'S limitation to no more than three. 3. It allows programmers to specify the lower bound of arrays, whereas in FORTRAN it was implicitly. 4. It supports nested IF statements, which was not the case in FORTRAN. 5. It has a "for" statement to do what FORTRAN'S "DO" statement does, but then it also has a "do statement to allow a sort of subprogram process, complete with parameters. 6. It formalized the concept of data type. 7. It added the idea of compound statements, which are program statements composed of multiple simple statements. 8. It introduced the concept of block structure. This allowed the programmer to localize parts of programs by introducing new data environments, or scopes. For example, in the following code segment begin integer total, index; end; the begin-end pair forms a block in which the variables total and index are local variables. They are not visible, nor do they have storage associated with them when control is outside the block. 9. It allows semi-dynamic arrays for which the subscript range or ranges are specified by variables, so that the size of the array is set at the time storage is allocated to the array which happens when the declaration is reached during execution. For example, the following code segment illustrates the use of this feature: procedure addelement (listlen, newelement); value listlen; integer listlen, newelement; begin integer array [1 : listlen] list; end end; addelement; ALGOL succeeded in becoming the only acceptable formal means of communicating algorithms, and it remained for over 20 years the sole language for publishing algorithms. However, ALGOL never became a popular programming language. Some of the main reasons for its lack of acceptance and wide-spread usage are:
1. Some of its features turned out to be too flexible, making its understanding difficult and implementation inefficient. 2. With its vast and flexible features, development of a compiler for the language became an extremely difficult task. 3. The language was primarily designed as a way of expressing algorithms, and input/output statements were not made a part of the language. The lack of I/O facilities made it very difficult to interface a program with the outside world. 4. IBM, the largest computer manufacturer during those days, did not favor and support ALGOL on its systems early in its existence. RPG RPG stands for Report Program Generator. As the name implies, it is a language designed to generate the output reports resulting from the processing of common business applications. The language was developed by IBM as a result of their customer requests for an easy and economic mechanism for producing reports. It was launched in the year 1961 for use on the IBM 1401 computer. The later version of RPG, called RPG II, greatly improved the language and gave it additional capabilities. RPG is considered different from other programming languages. Instead of writing instructions or statements, the programmer uses very detailed coding sheets to write his/her specifications about input, calculations, and output. These sheets specify exactly what is to be done, and then the computer uses them to generate the necessary instructions to perform the desired applications. Thus, RPG is easier to learn and use as compared to COBOL. Owing to these reasons, it is well suited for applications where large files are read, few calculations are performed, and output reports are created. However, RPG has restricted mathematical capability and cannot be used for scientific applications. RPG is primarily used to process business applications on small computers. LISP LISP stands for LISt Processing. It was developed in 1959 by John McCarthy of MIT. His goal was to develop a language that will be good at manipulating nonnumeric data, such as symbols and strings of text. Such data handling capability is needed in compiler development, and in Artificial Intelligence (AI) applications that deal with the idea of using computers for modeling human information storage and retrieval, along with other fundamental processes, of the brain. This capability was provided in the form of list processing feature that allow computers to process symbolic data in lists (collection of non-contiguous memory cells that are chained together with pointers). A list having four data elements is shown in Figure 12.23.
LISP is a functional programming language and is very different from other imperative languages like FORTRAN, COBOL, and Pascal. All computation in a functional programming language is accomplished by applying functions to arguments. There need not be any assignment statements or even variables that are used the way those of the imperative languages (to store values that change during program execution). Due to its powerful list processing capability, LISP is extensively used in the areas of pattern recognition, artificial intelligence, and for simulation of games. It totally dominated AI applications area for almost a quarter of a century. Even today, it is the most widely used language for AI applications. SNOBOL SNOBOL stands for StriNg Oriented symBOlic Language. It is another language used for non-numeric applications. As the name implies, the language was basically designed to manipulate strings of characters. It has powerful string manipulation features that facilitate various types of operations on strings of characters such as string comparisons, splitting of a string, combining two strings, etc. Thus, this language has been widely accepted for applications in the area of text processing. PROLOG PROLOG stands for PROgramming in LOGic. Logic programming deals with the use of a formal logic notation to communicate some computational process to a computer. Predicate calculus is the notation used in current logic programming languages. : PROLOG is a non-procedural language and is very different from other procedural languages like FORTRAN, COBOL, and Pascal. In a. procedural programming language, the programmer gives detailed instructions on how to do the job at hand. On the other hand, in a non-procedural programming language, the programmer only specifies what is to be done and leaves it to the system to figure out the methodology to do the job at hand. What is needed to provide this capability is a concise means of supplying the computer with both the relevant information and a method of inference for computing desirable results. PROLOG uses predicate calculus for supplying the basic form of communication to the computer, and the proof method named resolution, for the inference technique. Notice that a nonprocedural programming language allows the programmer to concentrate on the problem specification. Non-procedural programming languages are also known as problem-oriented or specification-oriented languages.
PROLOG was developed in early 1970s by a group consisting of scientists from the University of Aix-Marseille and the University of Edinburgh. Its first interpreter was developed at Marseille in 1972. It started with a lot of promise for use in the development of expert systems (intelligent computer systems). However, as of now it is not very widely used mainly because of the following two major limitations of logic programming languages: 1. As of now logic programming languages have proven to be highly inefficient. 2. It has been found that logic programming languages can be effectively utilized for only a few relatively small areas of application. CHARACTERISTICS OF A GOOD PROGRAMMING LANGUAGE In the previous section, we have seen that there are some high-level languages, which are very popular, and there are others, which could not become so popular in-spite of being very powerful. There might be many reasons for the success of a language, but one obvious reason is the characteristics of the language. Several characteristics believed to be important with respect to making a programming language good are briefly described below. 1. Simplicity. A good programming language must be simple and easy to learn and use. For example, BASIC is liked by many programmers only because of its simplicity. Thus, a good programming language should provide a programmer with a clear, simple, and unified set of concepts which can be easily grasped. It is also easy to develop and implement a compiler or an interpreter for a programming language that is simple. However, the power needed for the language should not be sacrificed for simplicity. The overall simplicity of a programming language strongly affects the readability of the programs written in that language, and programs that are easier to read and understand are also easier to maintain. 2. Naturalness. A good language should be natural for the application area it has been designed. That is, it should provide appropriate operators, data structures, control structures, and a natural syntax in order to facilitate the users to code their problem easily and efficiently. FORTRAN and COBOL are good examples of scientific and business languages respectively that posses high degree of naturalness. 3. Abstraction. Abstraction means the ability to define and then use complicated structures or operations in ways that allow many of the details to be ignored. The degree of abstraction allowed by a programming language directly affects its writability. For example, object-oriented languages support high degree of abstraction. Hence writing programs in object-oriented languages is much easier. Object-oriented languages also support reusability of program segments due to this feature. 4. Efficiency. The programs written in a good programming language are efficiently translated into machine code, are efficiently executed, and acquire as
little space in the memory as possible. That is, a good programming language is supported with a good language translator (a compiler or an interpreter) that gives due consideration to space and time efficiency. 5. Structuredness. Structuredness means that the language should have necessary features to allow its users to write their programs based on the concepts of structured programming. This property of a language greatly affects the ease with which a program may be written, tested, and maintained. Moreover, it forces a programmer to look at a problem in a logical way so that fewer errors are created while writing a program for the problem. 6. Compactness. In a good programming language, programmers should be able to express intended operations concisely. A verbose language can tax the programmer's sheer writing stamina and thus reduce its usefulness. COBOL is generally not liked by many programmers because it is verbose in nature and lacks compactness. 7. Locality. A good programming language should be such that while writing a program, a programmer need not jump around visually as the text of the program is prepared. This allows the programmer to concentrate almost solely on the part of the program around the statement currently being worked with. COBOL lacks locality because data definitions are separated from processing statements, perhaps by many pages of code. 8. Extensibility. A good programming language should also allow extension through simple, natural, and elegant mechanisms. Almost all languages provide subprogram definition mechanisms for this purpose, but there are some languages that are rather weak in this aspect. 9. Suitability to its Environment. Depending upon the type of application for which a programming language has been designed, the language must also be made suitable to its environment. Foi example, a language designed for real time applications must be interactive in nature. On the other hand, languages used for data processing jobs like pay-roll, stores accounting, etc., may be designed to be operative in batch mode. SELECTING A LANGUAGE FOR CODING AN APPLICATION One is often faced with the situation of selecting a programming language out of the many options available for coding an application. The following factors generally influence the selection process: 1. Nature of the application. The language should be suitable for the application area. For example FORTRAN is suitable for scientific and engineering applications, while COBOL is suitable for business applications. 2. Familiarity with the language. If there are multiple languages that are found suitable for the application area, then the language selected should be one that is best known to the programmers who are going to code the application.
3. Ease of learning the language. If there are multiple languages that are found suitable for the application area, and if the programmers are not familiar with any of them, then the language, which is easier to learn and use, should be selected. 4. Availability of program development tools. Before selecting a language, one must also find out whether the language is well supported with good program development tools like compiler, interpreter, debugger, linker, etc. The time and effort needed for coding the application can be greatly reduced if the selected language is supported with good program development tools. 5. Execution efficiency. If the execution efficiency of the application is important, one many use an assembly language instead of a high-level language for coding the application. This is because an assembly language program written by a clever programmer usually has a shorter production run time and takes less storage space than does a program of the same application written in a high-level language. 6. Features of a good programming language. Finally, the features of a good programming language discussed in the previous section often influence the selection process. OTHER RELATED CONCEPTS Subprogram Suppose we are writing a program for solving a trigonometric problem. In the problem solution, suppose we need to calculate the square root of three numbers at three different places in the program. We could of course write the steps required for the calculation of square root each time we need them in the program. But a much easier way of handling the problem would be to write these steps once and then refer to them each time we need them. Every programming language provides a way to do this by defining subprograms. A subprogram is a program written in such a way that it can be brought into use in other programs and used whenever needed, without rewriting. A subprogram is referred to by different names in different programming languages. For example, other names used to refer to subprograms are subroutines, subprocedures, and functions. The structure of a subprogram is shown in Figure 12.24. It consists of a header and a body. The subprogram header, which is the first line of a subprogram, provides a name for the subprogram and it may optionally specify a list of parameters (also known as arguments). The subprogram name is used by other programs to refer to the subprogram when they want to use the subprogram. The parameters are specific variable names, numeric values, or other data that you want the subprogram to use as it executes the instructions in its body. The list of parameters is optional because there may be subprograms for which you do not need or want
to pass any arguments. The subprogram body contains the set of instructions that perform the intended task of the subprogram. Once created, a subprogram can be used by other programs as often as needed. At whatever points in a program the function performed by the subprogram is required, a statement to call the subprogram is inserted by the programmer. Typically, the subprogram call statement contains the name of the subprogram followed by the list of parameters enclosed within a pair of parentheses. If you do not need or want to pass any arguments to the subprogram, the empty parentheses are still required normally to distinguish the statement as a subprogram call. For example, a statement to call the subprogram which calculates the square root of a number may be a = sqrt (b); Different programming languages may have different syntax for writing the call statement for calling a subprogram. Set of instructions that perform the intended task Subprogram header Subprogram body While executing a program that has one or more subprogram call statements, when the computer encounters a subprogram call statement, the control flow jumps to wherever the subprogram's code is, executes the statements in the subprogram body, then returns to that statement of the calling program which is immediately after the call statement. The flow of control in case of a subprogram call is illustrated in Figure 12.25. Note that the subprogram itself need never be written again, nor need it appear more than once in a program, no matter how often its services may be required. Subprograms may be intrinsic or programmer-written. Intrinsic subprograms (commonly referred to as built-in-functions) are those provided with the language so that the programmer need only call them in a program to have them automatically invoked. There are many functionalities such as those for finding square roots, sines, cosines, logarithms, etc., which are used over and over again by many programmers in their programs. Such functionalities are usually supplied along with a programming language in the form of built-in-functions. On the other hand, programmer-written subprograms are written and used as and when they are needed. Subprograms can also return a value to the calling program. This makes it convenient to use a subprogram to set the value of a variable. For example, in case of our subprogram to calculate the square root of a number, you supply the number as its parameter,, and the, subprogram calculates and returns the square root of the number. Based on this criterion, subprograms are often categorized into two basic types: procedures and functions. A procedure is a subprogram that
performs an action but returns no value. For example, a procedure might print a value or update a data structure. On the other hand, a function is a subprogram that returns a single value, as do mathematical functions like sine, cosine, absolute, square root. Both functions and procedures are widely used in general-purpose, high-level programming languages. Subprograms have the following distinct advantages: 1. They make it easier to write programs because the programmer can write less code. The same code can be called from many places in a program, instead of being duplicated everywhere it is needed. 2. Since only one copy of a subprogram's code is kept in the system instead of duplicating it everywhere it is needed, this leads to shorter source programs and their executable codes for programs that make use of subprograms. This leads to saving of storage space in the system and requirement of less memory space during the execution of a program. 3. Subprograms allow programs to be built in a modular fashion, which is essential to the construction of large software systems. 4. Modular design of software systems in the form of subprograms also provides the capability of compiling parts of a program (individual subprograms) without compiling the whole program (all subprograms used by the program). With such a capability, only the modules (subprograms) of a system that are being changed need to be recompiled during development or maintenance of the software system. Newly compiled and previously compiled modules (subprograms) are collected and put together by a program called linker. Without this capability, every change to a software system would require a complete recompilation. In a large system, this is costly. 5. Creating a subprogram for the task in question will obviate the need to do it again. This promise gives its author incentive to invest great care in generalizing it, debugging it, optimizing its performance, documenting it, and otherwise perfecting it. Thus subprograms often are very well written and documented parts of a program. Preprocessor A standard programming language is often extended to support features needed by a specific user group. For example, a user group frequently dealing with a special type of input data or a special type of input device may like to have statements to directly accept this input. In this case, a standard language, which is most suitable for the user group applications, can be extended to have statements to support this. With this extension, the users can program their applications more easily. However, since the extended features (additional statements) are not supported in the original standard language, a language processor is needed that translates the
additional statements (extensions to the standard language) into a sequence of statements of the standard language before the program can be successfully compiled by the standard language compiler. Such a language processor is known as a preprocessor. As shown in Figure 12.26, a preprocessor accepts as input a source program written in an extended language (with the added features) and produces an output source program that is acceptable to the compiler of the standard language (without the added features). The preprocessor reads each statement in the input source program and parses it to see if its syntax pertains to the added statements (extended features). If not, the preprocessor leaves the statement unchanged. If so, the preprocessor generates a sequence of statements in the standard language for the current statement 4o produce the same effect. Note that the output source program generated by the preprocessor will be longer than the input source program because the preprocessor generates a sequence of statements for each statement in the extended language. As shown in the figure, preprocessing inevitably requires two translation passes the first pass translates the source program in extended language to a standard language source program by using the preprocessor, and the second pass translates the standard language source program to executable object code by using the standard language compiler. But in most cases of preprocessing, the user is given the feel of one-pass translation through creation of a library of system commands that, when invoked, runs through the multiple operations in response to a short sequence of keystrokes or one click of a picking device. Scripting Languages Command statements are used to instruct a computer to do some job. For example, a "run" or "exec" statement may be used to execute a program and a "copy" statement may be used to copy a file from one storage device 1 another. Often a user needs to execute a set of commands to perform a particular task. For example, to perform task, the user may be required to execute two independent programs, and after execution of the first program he/she may be required to copy the output file of the first program on to a device from which the second program expects the input. So in this case, the user needs to specify two execute commands for executing the two programs and a copy command before the second execute command. A scripting language is a programming language that allows programmers to integrate several such commands into a single command so that the programmer need not type all the commands every time he/she needs to perform the intended task. A program written in a scripting language is commonly referred to as a script. Scripting languages support variables and simple control structures to allow programmers to write scripts, which can conditionally execute
or repeat the execution of certain commands. With these features, scripting languages may be used to define complex commands execution patterns. Another common use of scripting languages is in specifying execution environment setup for applications that are quite complex, yet flexible enough* to be "used in a variety of contexts. For an individual run of such an application, programmers must configure a setup to customize the application for its current run. This setup can be specified in the form of a script written in a scripting language. The script integrates individual commands into a single command that can be used to drive application execution. A scripting language is interpreted, rather than compiled. Hence a script can interact with a running program and change during the run. In addition to controlling a running program, scripts can preprocess and post process data. Scripting languages typically let programmers invoke individual application components directly. Application components are individual programs, such as data format converters and visualizers that carry out specific data processing tasks and can be used in different contexts . Some commonly known scripting languages are Unix Shell Script, Perl, Tel, and Python. A typical script looks like the one shown in Figure 12.27. users = 'who I we —1' if test $users -gt 10 then echo "More than 10 users logged on" fi Figure 12.27. A typical script. The first line of the script sets the variable 'users' equal to the value produced by the pipeline of commands 'who I we -F. This is in fact a count of the number of users currently logged on to the system. This value is then compared with the number 10, and if it is greater, then the message is displayed. Fourth-Generation Languages (4GLs) Some people in the computer field like to identify the major developments in programming language area by generations. According to them, machine languages, assembly languages, and high-level languages are identified as firstgeneration, second-generation, and third-generation languages respectively. Every new generation of language brought significant improvement in programmer productivity by making programming languages more and more user-friendly. Studies indicate that the move from first-generation machine languages to secondgeneration assembly languages brought approximately a seven-to-one improvement in programmer productivity. A much more significant improvement
in programmer productivity was found in the move from second-generation assembly languages to third-generation high-level languages. The efforts for the search of new programming languages which could further improve programmer productivity continued even after the advent of third-generation high-level languages. The result of these efforts has been the fourth-generation languages, which claim programmer productivity improvements of 200% to 1000% over third-generation high-level languages. There is a major shift in fourth-generation languages as compared to the languages of previous generations. All previous-generation languages were procedural languages whereas fourth-generation languages are nonprocedural languages. A procedural language (such as COBOL, FORTRAN, BASIC) is one that requires the programmer to spell out the steps in the processing procedures needed to achieve a desired result. On the other hand, a non-procedural language allows the user to simply specify "what" the output should be without describing all the details of "how" the data should be manipulated to produce that result. With this type of flexibility provided by fourth-generation languages, many computer users often find it easier and quicker to write a program themselves than to depend on a professional programmer who will write a program for generating the desired report. Fourth-generation languages are so user-friendly that with a day or two of training and practice, a computer-competent user can learn to write programs, make inquires, and generate the desired reports on his/her own. Fourth-generation languages use high-level English-like instructions. They require the programmer to specify only "what" to do, not "how" to do it. Most of the procedure portion (steps to carry out the intended task) of a fourth-generation language program is generated automatically by the language processor of the 4GL. Fourth-generation language features include limited mathematical capabilities, automatic report formatting, sequencing, and record selection by criteria. A typical fourth-generation language program looks like the one shown in Figure 12.28. The result of the execution of the fourth-generation language program of Figure 12.28 is shown in Figure 12.29. Some commonly known fourth-generation languages are ADR's Ideal, Software AG's Natural 2, Cinmon's Mantis, Mathematica Products Group's RAMIS II, and Information Builders' FOCUS. 1. 2. 3. 4.
FILE IS STUDENTSINFO LIST BY ROLLNO: NAME SEX PHY CHEM MATH SELECT CLASS = 8 TITLE: "MARKS OBTAINED BY THE STUDENTS OF CLASS 8"
5. COLUMN HEADINGS: "ROLL, NUMBER"; "STUDENT, NAME"; "SEX"; "MARKS IN, PHYSICS"; "MARKS IN, CHEMISTRY"; "MARKS IN, MATHEMATICS" Figure 12.28. A typical fourth-generation language program. MARKS
OBTAIN BY THE STUDENTS OF 8 ED CLASS ROLL STUDEN SEX MARKS MARKS IN MARKS IN T IN NUMBER NAME PHYSICS CHEMISTRY MATHEMATIC S 001 P. K. M 85 • 89 96 Sinha 002 N. C. Rao M 72 76 77 003 S. S. F 86 82 94 Pawar
Figure 12.29. Result of the execution of the fourth-generation language program of Figure 12.28 (only few records are shown here just to illustrate the nature of output). Points to Remember 1. A language that is acceptable to a computer system is called a computer language or programming language, and the process of writing instructions in such a language for an already planned program is called programming or coding. 2. All computer languages can be broadly classified into three categories, namely machine language, assembly language and high-level language. 3. A programming language that can be understood by a computer without using a translation program is called the machine language of the computer. It is normally written as strings of binary Is and Os. 4. A language which allows instructions and storage locations to be represented by letters and symbols instead of numbers is called an assembly language or symbolic language. A program written in an assembly language is called an assembly language program or a symbolic program. 5. An assembler is a translator program that translates (converts) an assembly language program (source program) into its equivalent machine language program (object program). 6. Machine and assembly languages are often referred to as low-level languages because they are machine dependent, they require the programmers to have a good knowledge of the internal structure of the computer being used, and they deal with machine-level coding requiring one instruction to be written for each machine-
level operation. High-level programming languages overcome these limitations of low-level programming languages. 7. A compiler is a translator program that translates (converts) a high-level language program (source program) into its equivalent machine-language program (object program). 8. In addition to translating high-level language instructions into machine language instructions, compilers also automatically detect and indicate certain types of errors in source programs. These errors are referred to as syntax errors. A compiler, however, cannot detect logic errors in source programs. 9. A linker is a software that takes multiple object program files (modules) of a software and fits them together to assemble them into the program's final executable form, which is sometimes called a load module. 10. An interpreter is a translator program that translates (converts) a high-level language program into its equivalent machine-language program. However, unlike a compiler which merely translates the entire source program into an object program and is not involved in its execution, an interpreter takes one statement of the source program, translates it into machine language instructions, and then immediately executes the resulting machine language instructions before taking the next statement for translation. No object program of the source program is generated by the interpreter. 11. Object-oriented programming (OOP) deals with solving problems by identifying the real-world objects of the problem and the processing required of those objects, and then creating simulations of those objects, their processes, and the required communications between the objects. 12. Conventional procedure-oriented programming languages use procedural abstractions that encapsulate sequences of operations into procedures (also known as functions, subroutines, or subprograms). On the other hand, object-oriented programming languages encapsulate data as well as sequences of operations into abstract entities called objects, providing a stronger encapsulation mechanism than procedure. Hence object-oriented programming provides a more powerful modeling tool than procedure-oriented programming. 13. Today many different high-level languages exist and are in use because each was designed for different specific purposes. Some of these are FORTRAN, COBOL, BASIC, Pascal, PL/1, C, C++, Java, Ada, ALGOL, RPG, LISP, SNOBOL and PROLOG. 14. Several characteristics believed to be important with respect to making a programming language good are simplicity, naturalness, abstraction, efficiency, structuredness, compactness, locality, extensibility, and suitability to its environment. 15. The factors that generally influence the process of selecting a programming language out of the many options available for coding an application are nature of the application, familiarity with the language, ease of learning the language,
availability of program development tools, execution efficiency, and features of a good programming language. 16. A subprogram is a program written in such a way that it can be brought into use in other programs and used whenever needed, without rewriting. Other names used to refer to subprograms are subroutines, subprocedures, and functions. 17. Subprograms are often categorized into two basic types: procedures and functions. A procedure is a subprogram that performs an action but returns no value, while a function is a subprogram that returns a single value. 18. Subprograms may be intrinsic or programmer-written. Intrinsic subprograms (commonly referred to as built-in-functions) are those provided with the language so that the programmer need only call them in a program to have them automatically invoked. On the other hand, programmer-written subprograms are written and used as and when they are needed. 19. A standard programming language is often extended to support features needed by a specific user group. However, since the extended features (additional statements) are not supported in the original standard language, a language processor is needed that translates the additional statements (extensions to the standard language) into a sequence of statements of the standard language before the program can be successfully compiled by the standard language compiler. Such a language processor is known as a.preprocessor. 20. Command statements are used to instruct a computer to do some job. A scripting language is a programming language that allows programmers to integrate several such commands into a single command so that the programmer need not type all the commands every time he/she needs to perform the intended task. A program written in a scripting language is commonly referred to as a script. 21. Fourth-generation languages (4GLs) are very high-level programming languages that use English-like instructions. They require the programmer to specify only "what" to do, not "how" to do it. Most of the procedure portion (steps to carry out the intended task) of a fourth-generation language program is generated automatically by the language processor of the 4GL. 22. A procedural language (such as COBOL, FORTRAN, BASIC) is one that requires the programmer to spell out the steps in processing procedures needed to achieve a desired result. On the other hand, a nonprocedural language (such as 4GLs) allows the user to simply specify "what" the output should be without describing all the details of "how" the data should be manipulated to produce that result. Questions 1. 2.
What is a programming language? Why is it so called? What is meant by coding a program?
3. Discuss the analogy between a computer language and a natural language. 4. How does a computer language differ from a natural language? 5. What is meant by syntax rules of a programming language? 6. Name the three categories of computer languages in order of their ease of usage. 7. What is a machine language? Why is it required? 8. What are the advantages and limitations of machine language? 9. When is a computer language called machine dependent? What is the main disadvantage of such a language? 10. A machine language instruction has two-part format. Identify these parts and discuss the function of each. 11. Explain why machine language programming is difficult and error-prone. 12. It is said that machine language programming is difficult and error-prone. Explain how assembly languages overcome these limitations of machine languages. 13. What is a mnemonic? How is it helpful in case of computer languages? 14. What is an assembly language? What are its advantages over machine language? 15. What is an assembler? 16. What is the difference between a source program and an object program? 17. List out the advantages of assembly language over machine language. 18. List out the limitations of machine language which are not overcome by the use of assembly language. 19. List out some of the typical uses of assembly language. 20. What is a macro instruction? How does it help in reducing a programmer's job? 21. Why are machine and assembly languages often referred to as low-level languages? 22. List out the main characteristic features of high-level languages. Name ten high-level languages. 23. Why are high-level languages easier to learn and use? 24. What is a compiler? Why is it required? A computer supports five highlevel languages. How many compilers will this computer have? 25. It is said that an assembly language is "one-for-one" but a high-level language is "many-for-one". Explain what this means. 26. Illustrate the machine independence characteristic of a high-level language. 27. Differentiate between syntax errors and logic errors found in computer programs. Which type of error more difficult to detectand correct? Give reasons for your answer. 28. While writing a program, a programmer erroneously instructed the computer to calculate the area of rectangle by adding the width to its length (that is, AREA = LENGTH + WIDTH) instead of multiplying the length and the width. Would you expect the language processor to detect this error? Explain why.
29. A programmer eliminates all language processor errors from his/her program and then runs it to get print results. The programmer therefore concludes that the program is complete. Comment. 30. What type of errors in a program can be detected by a compiler? What type of errors cannot be detected? 31. What is a linker? Why is it required? 32. Differentiate between the compilation and linking processes of a software. 33. Differentiate among source programs, object programs, library modules, and load module of a software; system. 34. Explain the differences among assemblers, compilers, interpreters, and linkers. 35. What is an interpreter? How does it differ from a compiler? 36. What are the two approaches commonly used to translate high-level languages into machine language Discuss the relative advantages and disadvantages of the two approaches. 37. What advantages do compilers have over interpreters? 38. What advantages do interpreters have over compilers? 39. When would you prefer to use a compiler than an interpreter? 40. When would you prefer to use an interpreter than a compiler? 41. What are the advantages and limitations of high-level languages as compared to machine and assembly languages? 42. What is object-oriented programming? 43. It is said that the most essential part of support for object-oriented programming is the object mechanism What are the basic concepts used in object-oriented programming languages to support the object mechanism? 44. With reference to object-oriented programming languages define the following terms: (a) Object (e) Message protocol (b) Method (f) Multiple inheritance (c) Class (g) Instance of a class (d) Inheritance 45. Give a suitable example to illustrate the concepts of object, class and inheritance used in object-oriented programming languages. 46. Differentiate between procedure-oriented and object-oriented programming paradigms. 47. List out the advantages of object-oriented programming paradigm over procedure-oriented programming paradigm. 48. What is polymorphism in the context of programming languages? Explain with an example. 49. Since high-level languages are machine independent, one good high-level language should have been sufficient for use on all computer systems. In-spite of this fact, why are there so many high-level languages in use today?
50. Briefly describe the development and use of the following programming languages: (a) FORTRAN (h) Ada (b) COBOL (i) RPG (c) BASIC (j) ALGOL (d) Pascal (k) SNOBOL (e) C (1) LISP (f) C++ (m) PROLOG (g) PL/1 51. What is meant by standardization of a language? Why is it important? 52. What are comment statements? Why are they used in a program? How are they treated by a language processor? 53. Name the four divisions of a COBOL program. What is the purpose of each division? 54. What is a self-documenting language? Illustrate with an example? 55. Why is COBOL often referred to as a self-documenting language? 56. The main drawback of COBOL is that it is a verbose language. How/has this drawback been taken care of now to a large extent? 57. List out some of the program preparation techniques that are often included under the term "structured programming". 58. In Java, the compiled code is machine independent. Explain how is this achieved. 59. What is HotJava? 60. What characteristics are desirable for a good programming language? 61. What factors generally influence the process of selecting a programming language out of the many options l available for coding an application? 62. Would you be equally likely to choose FORTRAN or COBOL for programming a given problem? Why? 63. What is a subprogram? How do subprograms help in program writing? 64. Subprograms are often categorized into two basic types: procedures and functions. Differentiate between the two types. 65. What is a built-in-function? Give few examples of built-in-functions. 66. Illustrate with an example the flow of control in case of subprogram calls. 67. What is a preprocessor? Explain how a source program written in an extended language is converted into its equivalent executable object code. 68. What is a scripting language? What is a script? Explain with an example. 69. What are fourth-generation languages? How are they different from conventional high-level languages? 70. Differentiate between a procedural language and a non-procedural language. 71. Give the full form of the following: (a) FORTRAN (f) ALGOL (b) COBOL (g) LISP (c) BASIC (h) SNOBOL
(d) PL/1 (e) RPG
(i) OOP (j) 4GL
Chapter 13 System Implementation and Operation In the previous two chapters, we have discussed the analysis, design, and coding phases of software development. After the computer programs have been prepared, the developed software along with the associated hardware enters the implementation and operation phase. The goal of this chapter is to describe the main activities of the implementation and operation phase, which include testing and debugging of programs, complete documentation of the system, changeover to the new system, and system modification and maintenance. TESTING AND DEBUGGING Definition of Testing and Debugging So long as computers are programmed by human beings, computer programs will be subject to errors. Program errors are known as bugs, and the process of detecting and correcting these errors is called debugging. In general, testing is the process of making sure that the program performs the intended task, and debugging is the process of locating and eliminating program errors. Testing and debugging are important steps in program development. They are also timeconsuming steps. In fact, for large programs, the time spent in testing and debugging often exceeds the time spent in coding the program. Types of Program Errors In general, there are two types of errors that occur in a computer program - syntax errors and logic errors. They are briefly described below. Syntax Errors Syntax errors result when the rules or syntax of the programming language are not followed. Such program errors typically involve incorrect punctuation, incorrect word sequence, undefined terms, or misuse of terms. For example, the FORTRAN statement C = (A + B/2 has a syntax error because of the missing closing parenthesis which should be placed in the appropriate place depending on the logic of the program. All syntax errors in a program must be found and corrected before it will run. That is, a program cannot be successfully compiled and executed until all its syntax errors have been corrected. Logic Errors
A logic error is an error in planning the program's logic. Such errors cause the program to produce incorrect output. That is, a program which is free of syntax errors but has one or more logic errors will be successfully compiled and executed but will not produce correct output. The problem is that since the program contains logic errors, the logic being followed during its execution will not be correct and hence the program will not produce the desired results. For example, if a FORTRAN instruction should be "A = B + C" but has been coded as "A = E - C", it is an example of a logic error because the output produced by the program will not be correct. Testing a Program In case of a logic error, since no language rules have been violated, the computer does not know that an error ha been made. Hence it will not produce any error message for a logic error. Thus identifying logic errors in program is the responsibility of the program development and testing team. The team tests the program to determine whether or not it contains any logic error. The testing procedure involves running the program to process input test data and comparing the produced results with the known correct results. If the results generate for the test data do not match with the known correct results, it is assumed that the program contains one or mo: logic errors. Logic errors are typically due either to missing logic or to incorrect logic. If the logic is incorrect, the result generated from the test data will be wrong. These errors are the easiest of the logic errors to find. Errors caused by missing logic result from logical situations that the program was not designed to handle. As an example suppose that a numeric field is to be used in an arithmetic process and that the data entry operator enters a value for the field that is not numeric. The program logic should determine that the data are not numeric prior attempting the arithmetic process. If that logic is missing and nonnumeric data is used in an arithmetic operation, the program will fail. This type of logic error can be difficult to find. The only way for this error to occur is for nonnumeric data to be entered into a numeric field. It is possible for the program to be used for weeks, months, or years before this happens and the error in program logic shows up. In order to completely test the program logic, the test data must test each logical function of the program. Hence, the selection of proper test data is important in program testing. In general, the test data selected for testing a program should include: 1. All types of possible valid data. This will test the generally used program paths and will ensure that the program has been correctly designed to handle all types of input data that falls within the scope of the program.
2. All types of possible invalid data (such as incorrect, incomplete, or inappropriate data). This is necessary to test the error-handling capabilities of the program (how the program reacts in case of abnormal and unusual input data). Good programs should be able to produce appropriate error messages instead of hanging (not continuing with its execution) or generating meaningless output. If a program runs successfully with the test data and produces correct results, it is normally released for use. However, even at this stage errors may still remain in the program. In case of a complex program, there may be thousands of different possible paths through the program and it may not be practical or even possible to trace through all these paths during testing. There are certain errors in complex software systems that remain dormant for months and years together and hence it is impossible to certify that such systems are free of all logic errors.. Hence if a software is going to be sold commercially, instead of releasing it directly in the commercial market, it is first released to a selected set of users. This release version of the software is known as beta version. The selected set of users use the beta version of the software for an identified period of time and report any bugs found in the software to the company that is developing the software. The use of beta version of the software by the selected set of users for identifying errors that might have been overlooked during the regular testing phase is known as beta testing. After beta testing, the reported bugs (if any) are fixed and then the software is released to the commercial market. Note that even at this stage, the software may contain certain errors. Debugging a Program for Syntax Errors Syntax errors and logic errors are collectively known as bugs and the process of eliminating these errors is known as debugging. As compared to logic errors, syntax errors are much easier to eliminate because almost all language processors are designed to detect syntax errors. The language processors print error messages on the source listing of a program that indicate each program statement with errors and give hints as to the nature of the error. These error messages are very useful and are used by the programmers to rectify all syntax errors in their programs. Thus, it is a relatively easy task to detect and correct syntax errors. It should be noted that in high-level languages, such as FORTRAN and COBOL, a single error often causes multiple error messages to be generated. There are two reasons for multiple error messages. One is that high-level language instructions often require multiple machine steps. The other reason is that symbolic instructions are often dependent upon other instructions and if an instruction containing an error is one that defines a field name, all instructions in the program using that field name will be listed as errors. The error message will say that a field being used is not a defined name. In such a case, removal of the single error will result in the removal of all associated error messages.
Debugging a Program for Logic Errors Unlike syntax errors, the computer does not produce any error message for logic errors in a program. Hence logic errors are much more difficult to eliminate than syntax errors. However, once the testing of a program indicates the presence of logic error in it, one or more of the following methods may be used to locate and correct logic errors. Doing Hand Simulation of the Program Code One approach is to take a printout of the source code of the program and go through its execution manually with the test data input that produced incorrect results. In the manual execution, you follow the same sequence of paths, which the computer would have followed for the test data keeping track of the changes in the values of the various variables that you come across during the execution. The cause of the problem is very likely to be detected before you complete the manual execution. Manual execution of a program is a tedious process requiring lot of concentration. Hence this method is normally used only for simple and small programs. Putting Print Statements in the Program Code Another approach is to put several print or write statements at appropriate locations in the program so that the values of different variables can be printed/displayed to indicate intermediate computations results. The program is recompiled and executed with these statements. These statements print/display the values of intermediate computations telling you what is happening during the execution of the program. The cause of the problem is very likely to be detected by doing a proper analysis of the printed/displayed intermediate results. Once the errors have been found and corrected, these print or write statements are removed from the program. Using a Debugger This is the most commonly used approach. A debugger is a software tool that assists the programmer in following the program's execution step-by-step by allowing him/her to display intermediate calculation results and field values whenever desired. In effect, a debugger helps the programmer in debugging the program logic much more easily than having to put (and later remove) print/write statements into the program's source code. Almost all modern programming language processors are supported with some kind of debugger.
When using a debugger for debugging program logic, you can move either from statement to statement or from one checkpoint to another, pausing the execution of the program as needed to examine the values of key variables or machine registers. Most debuggers support two kinds of checkpoints called breakpoints and watchpoints. They are briefly explained below: 1. Breakpoint. A breakpoint is a point at which the debugger stops during program execution and awaits your further command. When it stops you may instruct it to display the current values of key variables or machine registers. A debugger provides you the flexibility of setting or removing breakpoints after any program statement. You may even set multiple breakpoints, and set or remove breakpoints when the program execution stops and waits for your further command at a breakpoint. 2. Watchpoint. Suppose a program contains a variable named "radius" whose initial value of 5 is not supposed to change during execution of the program or after some point in the program. But during debugging of the program you discover that its value is changing and you want to know the exact point in the program where its value gets changed. In such a situation, you can tell the debugger that "radius" is a watch variable. That is, you are telling the debugger to keep track of the variable "radius" and the moment its value changes, stop and give me control. The debugger will immediately stop after the statement that causes the value of the variable "radius" to change and will await your further command. Both kinds of checkpoints are very helpful in detecting logic errors in programs. Using Memory Dump This approach is normally used when the program "hangs up" during a test run. In this approach, a printout of the contents of main memory and registers is taken at the time when the program hangs up. This printout is called a memory dump or storage dump. The memory dump lists the instructions and data held in the computer's main memory in their raw form, that is, their binary or equivalent hexadecimal or octal form. The cause of the problem is very likely to be detected by doing a proper analysis of this listing. Difference Between Testing and Debugging Since both testing and debugging deal with program errors, many people often confuse between these two terms. The discussion above has clearly indicated that testing and debugging are two separate tasks. The points listed in Figure 13.1 further clarify the difference between the two. SI. Testing Debugging No.
1
2
Testing is the process of validating the correctness of a program. Its objective is to demonstrate that the program meets its design specifications. Testing is complete when all desired verifications against specifications have been performed.
3
Debugging is the process of eliminating errors in a program. Its' objective is to detect the exact cause of and remove known errors in the program. Debugging is complete when all known errors in the program have been fixed. Note that debugging process ends only temporarily because it must be restarted whenever a new error is found in the program. Debugging being a reactive process cannot be planned ahead of time. It must be carried out as and when errors are found in a program. Debugging can begin only after the program is coded. The approach used for debugging largely depends on the personal choice of the programmer and the type of problem in the program.
Testing is a definable process which can and should be planned and scheduled properly. 4 Testing can begin in the early stages of software development. Although the test runs of a program can be done only after the program is coded, but the decision of what to test, how to test, and with what kind of data to test, can and should be done before the coding is started. Figure 13.1. Difference between testing and debugging. DOCUMENTATION
What is Documentation? Documentation is the process of collecting, organizing, storing, and otherwise maintaining a complete historical record of programs and other documents used or prepared during the different phases of the life cycle of a software. A software cannot be considered to be complete until it is properly documented. In fact, documentation is an on-going process that starts in the study phase of the software and continues till its implementation and operation phase. Moreover, documentation is a process that never ends throughout the life of the software. It has to be carried out from time to time as and when the software is modified during its maintenance phase. Need for Documentation
Proper documentation of a software is necessary due to the following reasons: 1. It solves the problem of indispensability of an individual for an organization. In large organizations, over the years, the designer and developer of a software may not be in the same organization. In such cases, even if the person who has designed or programmed the software, leaves the organization, the documented knowledge remains with the organization, which can be used for the continuity of the software. 2. Maintainability of computer software is a cause of major concern to organizations dealing with computerized systems. The key to maintenance is proper and dynamic documentation. A well documented software is easy to modify and maintain in the future. It is easier to understand the logic of a program from the documented records rather than its code. System flowcharts, program flowcharts, or comments used within the programs prove to be very helpful in this regard. 3. Documented records are quite helpful in restarting a software project that was postponed due to some reason or the other. The job need not be started from scratch and the old ideas may still be easily recapitulated which saves lot of time and avoids duplication of work. Forms of Documentation The three commonly used forms of documentation for documenting a software are comments, system manual and user manual. They are briefly described below. Note that all three forms of documentation are necessary to properly and completely document a software. Comments Comments are very useful aid in documenting a program. From maintenance point of view, comments have been considered to be a must. Comments are used within a program to assist anyone reading the source program listing. They are used to explain the logic of the program. They do not contain any program logic and are ignored (not translated) by the language processor. Comments should be used intelligently to improve the quality and understandability of the program. They should not be redundant, incorrect, incomplete, or written in such a way that cannot be understood by anyone else. For example, a redundant comment for the statement N = N + 1 would be "INCREMENT N BY 1". Useful comments are those which describe the meaning of a group of statements such as "READ. AND ECHO PRINT THE INPUT DATA". In other words, comments should mediate between the program and the problem domain.
Almost all high-level languages provide the facility of writing comments along with the source code of a program and it is suggested that programmers should liberally use this facility for proper documentation of their programs. System Manual A good software must be supported with a standard system manual that contains the following information: 1. A statement of the problem clearly defining the objectives of developing the software and its usefulness to various categories of users. 2. A broad description of the software specifying the scope of the problem, the environment in which it functions, its limitations, the form and type of input data to be used, and the form and type of output required. 3. Specific program names along with their description and purpose. 4. Detailed system flow charts and program flow charts cross-referenced to the program listing. 5. Narrative description of the program listings, the processing and calculations performed and the control procedures. 6. A source listing of all the programs together with full details of any modifications made since. 7. Description and specifications of all input and output media required for the operation of various programs. 8. Specimen of all input forms and printed outputs. 9. File layout, that is, the detailed layout of input and output records. 10. The structure and description of test data and test results, storage dumps, trace program printouts, etc., used to debug the programs. User Manual A good software package must be supported with a good user manual to ensure the smooth running of the package. It is the user who will perform the regular processing after the software is released and not the programmer who has developed the package. Hence, the user manual must contain the following information: 1. Set up and operational details of each program. 2. Loading and unloading procedures. 3. Starting, running, and terminating procedures. 4. A description and example, of any control statements that may be used. 5. All console commands along with errors and console messages that could arise, their meaning, reply and/or operation action. 6. List of error conditions with explanations for their re-entry into the system.
7. List of programs to be executed before and after execution of each program.384 I Foundations of Computing 8. Special checks (if any) and security measures, etc. Documentation Standard The importance of software documentation cannot be over emphasized. There have been too many problems in the past with poorly documented software. The result is usually errors and problems with the software at a later date. It is very difficult to incorporate modifications in such software and hence they are very expensive to maintain. Owing to these reasons several computer organizations have developed strict documentation standards. These standards describe in detail how documentation is to be performed, how to choose meaningful program variable names, how to design the GUI for the software, how and up to what detail to include comments in program code, and what diagrams, charts, reports, outputs, etc. are necessary for documentation to be completed successfully. CHANGEOVER TO THE NEW SYSTEM Once the software is thoroughly tested and all known errors have been removed, the software along with the associated hardware is deployed at site for use by the intended users. At this stage, the old system (manual system in most cases), if any, is phased out and the new system is phased in. Changeover Operations The changeover process normally involves the following operations: 1. Imparting system and user training to the members of the organization in which the new system is deployed. The system training is imparted to those members who will be responsible for managing and maintaining the system. The user training is imparted to those members who will be operating the system and/or using the results produced by the system. Anyone else who will be affected by the new system should also receive some training to become familiar with the changes. 2. Replacing all old operation procedures by new ones. This may involve discarding of old forms, manual registers, etc. 3. Replacing of all old input and output devices with those of the new system. 4. Defining the roles of different members and assigning the responsibilities to them as per the requirements of the new system. 5. Converting data in all currently existing files into a form acceptable to the new system. This may involve inputting of data stored in manual registers and hard-copy files through the input devices of the new system. It is important to consolidate the files and eliminate duplicate records in them while converting
them to the new form. File inconsistencies or any errors in existing files must also be detected and removed. Changeover Methods The three normally followed methods to carry out the changeover process are immediate changeover, parallel run and phased conversion. Each of these methods has its own advantages and disadvantages. No single method is suitable for converting all types of systems and the choice of a particular method largely depends upon the prevailing conversion circumstances. These methods are briefly described below. Immediate Changeover As shown in Figure 13.2(a), in this method, a suitable cut-off data is decided and the complete new system is put to operation from that day onwards, while the operation of the old system is completely abandoned from the cutoff date. It has been found that most systems pose some problem or the other during the changeover process. Hence the method of immediate changeover is generally considered to be risky because any failure in the new system during the changeover phase may cause total breakdown of those operations of the organization that are related to the new and the old system. The work cannot progress at all because the operation of the old system has already been stopped. This method, however, is preferred in those situations where available manpower and changeover time is less, and the system is not so critical that some operational problems during the changeover process would lead to a disaster. Parallel Run As shown in Figure 13.2(b), in this method, both the old and the new systems are parallely operated with the same data for the initial three or four cycles. During this overlapping period of complete operation of the two systems, the results produced by the new system are compared with those produced by the old system to develop confidence in the new system. Some discrepancies may be discovered. Often these are due to inaccuracies in the old system that were not recognized before as inaccuracies. Some discrepancies may also be due to missing program logic for which no programming was provided or due to mistakes in the programming itself. These must be corrected by further debugging before the conversion is complete.
This method is considered to be one of the safest ways to deploy a new system because of the availability of old system as a backup. There is no interruption of work if there are problems with the new system because the old system is still in operation and the problems found in the new system can be corrected while the old system is still being used. This method, however, is very expensive because additional manpower is needed during the overlapping period for the operation of two systems in parallel. Due to the requirement of additional manpower, the organization is under considerable strain during the period of parallel operation, and organizational breakdowns tend to occur if the period of parallel operation is long. Hence parallel operation must not be carried on any longer than needed to establish confidence in the new system. Continuing them too long is a sign of weakness in the new system. Due to the reason mentioned above, this method is not preferred in situations where manpower resource is scarce. It is also not used in situations where the new system differs to a great extent from the old system in the functions that it performs and its input and output. Phased Conversion As shown in Figure 13.2(c), in this method, the complete changeover to the new system takes place incrementally over a period of time. The new system is gradually implemented part by part and the old system is gradually phased out. The results produced by each part of the new system are compared against the results of the old system. Any discrepancies or errors found are checked and removed. Once confidence is developed in a particular part of the new system, that part of the new system is phased in and the corresponding part (operations) of the old system is phased out. This approach is continued for each and every part of the new system. Thus, over a period of time, the new system is gradually phased in, while the old system is gradually phased out. This method is not as expensive as the parallel run method because the changeover process being gradual can usually be handled with existing manpower. Moreover, there is no danger of interruption of work if there are problems with the new system because the corresponding part of the old system is still in operation. The users also get sufficient time to become acquainted with the new system. Hence, they can confidently handle the new system when the complete system is handed over to them. This method, however, cannot be used in situations where the time period supplied for conversion process is very less or when the new system significantly differs from the old system.
SYSTEM EVALUATION Once the new system is implemented and put in operation, it is necessary to evaluate the system to verify whether or not it is meeting its objectives. These objectives of the system are clearly stated during its problem analysis and system planning phase. The post implementation system evaluation is normally carried out by people who have an independent view point and are not responsible for the development and maintenance of the system. While evaluating a system, the following points are normally considered: 1. Performance Evaluation. The performance of the new system is evaluated and compared with the performance of the old system. In general, the new system should be at least as efficient as the old one in performance. In case of any slack, the reason is analyzed and if possible, necessary changes are incorporated in the new system to rectify it. 2. Cost Analysis. It should be analyzed whether the cost estimate done for the various phases of the project during the planning phase in the beginning match with the actual cost incurred in each phase or not. If discrepancies are found, the cost analysis will help in finding out the reasons. This type of analysis proves to be quite helpful in making correct cost estimates for the new systems that will be designed in future. 3. Time Analysis. It should be analyzed whether the time estimate done for the various phases of the project during the planning phase in the beginning match with the actual time taken in each phase or not. If discrepancies are found, the time analysis will help in finding out the reasons. This type of analysis proves to be quite helpful in making correct time estimates for the new systems that will be designed in future. 4. User Satisfaction. People are the final evaluators of any system put in operation. Hence, it should be found out whether the users are satisfied with the new system or not. How useful is the system for them? How enthusiastic are they about the service they receive? Do they receive outputs in time to take necessary action? The morale of people using or affected by a system is a good measure of the success of the system. 5. Ease of Modification. Sooner or later, all systems need to be modified due to one or more of the reasons mentioned in the next section. Thus, the ease with which a system can be modified to incorporate the suggested changes is also an important parameter to judge the quality of the system. 6. Failure Rate. Another parameter that is commonly used to judge the quality of a system is its failure rate. A system, which frequently fails cannot meet its objectives successfully and hence is of poor quality. SYSTEM MAINTENANCE
No matter how good the new system is, how well it was installed, and how well it may be currently operating sooner or later it will need to be modified due to one or more of the following reasons: 1. Changes in business conditions or operations of the organization using the system. 2. Changes in organizational policies or enforcement of new laws. 3. Changes in user needs such as demand for additional features, or additional reports and outputs from the system. 4. Changes in technology such as availability of new and more powerful processors, more efficient storage, more convenient input/output devices, new system software, etc. System maintenance is an important phase of the overall life cycle of a computerized system and its importance should not be under estimated. In fact, several studies have shown that, on an average, the maintenance cost of a computerized system is two to four times more than the initial development cost. This means that more than half of the programming tasks in this world involve program maintenance. Due to this reason, in some organization there are programmers who do nothing but maintain production programs. Frequent change is disrupting and disturbing. Therefore, some control over changes is required. One method o achieving this control is to have all requests for change evaluated by a change control board. This board should be made up of the principal users of the system, a system analyst, and data processing personnel who are familiar with the system. Normal maintenance operations need not be approved by the change control board, but these operations should be recorded and summarized for periodic reporting to the board. Examples of maintenance activities are modifying the format of a report or rewriting a part of a computer program component to improve ii efficiency. Major changes are those that significantly alter the system or require extensive personnel, hardware or software. An example of a major change would be conversion of a system from Unix to Windows environment. When programs are modified, it is important to make sure that program documentation is also changed accordingly. Without the existence of proper documentation that is consistent with the programs, future changes would be very difficult and costly to accomplish. Points to Remember 1. Testing is the process of making sure that the program performs the intended task. 2. Program errors are known as bugs, and the process of detecting and correcting these errors is called debugging. 3. There are two types of errors that occur in a computer program - syntax errors and logic errors. Syntax errors result when the rules or syntax of the programming
language are not followed. Logic errors are errors in planning the program's logic. Such errors cause the program to produce incorrect output. 4. As compared to logic errors, it is relatively easier to detect and correct syntax errors because the language processors print error messages on the source listing of a program that indicate each program statement with errors and give hints as to the nature of the error. 5. The commonly used methods for locating and correcting logic errors are doing hand simulation of the program code, putting print statements in the program code, using a debugger, and using memory dump. 6. A commercial software is first released to a selected set of users. This release version of the software is known as beta version. The use of beta version of the software by the selected set of users for identifying errors that might have been overlooked during the regular testing phase is known as beta testing. 7. A debugger is a software tool that assists the programmer in following the program's execution step-by-step by allowing him/her to display intermediate calculation results and field values whenever desired. 8. When a program hangs up during a test run, a printout of the contents of main memory and registers is taken to analyze the cause of the problem. This printout is called a memory dump or storage dump. 9. Documentation is the process of collecting, organizing, storing, and otherwise maintaining a complete historical record of programs and other documents used or prepared during the different phases of the life cycle of a software. The three commonly used forms of documentation for documenting a software are program comments, system manual and user manual. 10. Once the software is thoroughly tested and all known errors have been removed, the software along with the associated hardware is deployed at site for use by the intended users. At this stage, a changeover from the old system to the new system takes place. The three normally followed methods to carry out the changeover process are immediate changeover, parallel run and phased conversion. Figure 13.2 illustrates the three methods. 11. System evaluation is the process of evaluating a system (after it is put in operation) to verify whether or not it is meeting its objectives. The points normally considered for evaluating a system are performance evaluation, cost analysis, time analysis, user satisfaction, ease of modification and failure rate. 12. System maintenance is the process of incorporating changes in an existing system to enhance, update or upgrade its features. On an average, the maintenance cost of a computerized system is two to four times more than the initial development cost. Questions 1. What are the two types of errors that can occur in a computer program? Give an example of each to illustrate their nature.
2. Differentiate between syntax errors and logic errors in computer programs. 3. Will a program run if there are syntax errors in it? Give reasons for your answer. 4. Will a program run if there are logic errors in it? Give reasons for your answer. 5. What is testing of a program? Why should a program be tested? 6. What are the different types of test data that should be selected for testing a program? 7. Why is it not possible for a very complex software system to certify that it is error free? 8. What is a beta version of a commercial software? What is beta testing? 9. How are syntax errors detected and corrected? 10. How are logic errors detected and corrected? 11. Is it easier to detect a syntax error or a logic error in a program? Give reasons for your answer. 12. What is a program bug? What is debugging? 13. In high-level languages, a single syntax error often causes multiple error messages to be generated. Explain why. 14. Describe the various methods that may be used to locate and correct logic errors. 15. What is a debugger? How does it help a programmer? 16. In the context of debugging of a program, what is a checkpoint? What are the two commonly supported types of checkpoints in modern debuggers? How are they useful? 17. Differentiate between a breakpoint and a watchpoint in the context of program debugging. 18. What is a memory dump? How is it useful for a programmer? 19. Differentiate between testing and debugging. 20. Testing is a definable process, which can and should be planned and scheduled properly. Discuss. 21. Testing can begin in the early stages of software development while debugging can begin only after the program is coded. Discuss. 22. What is software documentation? Why is it needed? 23. Describe the different types of documentation normally used for documenting a software system. 24. What are program comments? Why are they needed? 25. What is a system manual of a software system? What type of information does it typically contain? 26. What is a user manual of a software system? What type of information does it typically contain? 27. What is documentation standard? Why is it needed? 28. What type of operations is normally carried out in the system changeover process?
29. Describe the three different methods of system changeover along with their advantages and disadvantages. 30. Immediate changeover is a risky method of deploying a new system. Explain why. 32. In what situations is the parallel run method of deploying a new system not preferred? 33. Why is system evaluation needed? What are the main parameters normally used to evaluate a new system? 34. Why do we need to modify a system if it is currently operating well? 35. Why is system maintenance considered an important process in the overall life cycle of a computerized system? 36. How can frequent program modifications be controlled?
Chapter 14 Operating Systems In the last few chapters we have dealt with the planning, coding, operation, and maintenance of software systems. In the next few chapters, we will have a look at some of the commonly used software in today's computer systems. In this chapter, you will learn about a very important and special type of software that falls under the category of system software. This system software is known as operating system. In the next chapter you will learn about some commonly used application software packages. An operating system is an important part of almost every computer system. The goal of this chapter is to introduce concepts related to operating systems and to show how this particular system software helps in the overall operation and usage of a computer system. WHAT IS AN OPERATING SYSTEM? An operating system (often referred to as OS) is an integrated set of programs that controls the resources (the CPU, memory, I/O devices, etc.) of a computer system and provides its users with an interface or virtual machine that is more convenient to use than the bare machine. According to this definition, the two primary objectives ol an operating system are as follows: 1. Making a computer system convenient to use. A computer system consists of one or more processors, main memory and many types of I/O devices such as disks, tapes, terminals, network interfaces, etc. Writing programs for using these hardware resources correctly and efficiently is an extremely difficult job, requiring in-depth knowledge of the functioning of the resources. Hence to make computer systems usable by a large number of users, it became clear several years ago that some way is required to shield programmers from the complexity of the hardware resources. The gradually evolved solution to handle this problem is to put a layer of software on top of the bare hardware, to manage all parts of the system, and present the user with an interface or virtual machine that is easier to program and use. This layer of software is the operating system. The logical architecture of a computer system is shown in Figure 14.1. As shown in the figure, the hardware resources are surrounded by the operating system layer, which in turn is surrounded by a layer of other system software (such as compilers, editors, command interpreter, utilities, etc.) and a set of application programs (such as commercial data processing applications, scientific and engineering applications, entertainment and educational applications, etc.).
Finally, the end users view the computer system in terms of the user interfaces provided by the application programs. Users Other System Software and Application Programs Operating System Computer Hardware The operating system layer provides various facilities and services that make the use of the hardware resources convenient, efficient, and safe. A programmer makes use of these facilities in developing an application, and the application, while it is running, invokes the required services to perform certain functions. In effect, the operating system hides the details of the hardware from the programmer and provides the programmer with a convenient interface for using the system. It acts as an intermediary between the hardware and its users, providing a high-level interface to low-level hardware resources, and making it easier for the programmer and for application programs to access and use those resources. 2. Managing the resources of a computer system. The second important objective of an operating system is to manage the various resources of the computer system. This involves performing such tasks as keeping track of who is using which resource, granting resource requests, accounting for resource usage, and mediating conflicting requests from different programs and users. Executing a job on a computer system often requires several of its resources such as CPU time, memory space, file storage space, I/O devices, and so on. The operating system acts as the manager of the various resources of a computer system and allocates them to specific programs and users to execute their jobs successfully. When a computer system is used to simultaneously handle several applications, there may be many, possibly conflicting, requests for resources. In such a situation, the operating system must decide which requests are allocated resources to operate the computer system efficiently and fairly (providing due attention to all users). The efficient and fair sharing of resources among users and/or programs is a key goal of most operating systems. Main Functions of an Operating System We saw that an operating system provides certain services to programs and to the users of those programs. The; specific services provided will, of course, differ from one operating system to another, but there are some common types of functions that we can identify. The main functions provided by most operating systems of today are as follows:
1. Process Management. A process is a program in execution. During execution, a process needs certain resources such as CPU time, memory space, files, and I/O devices. At a particular instance of time, a computer "system normally consists of a collection of processes. The process management module of an operating system takes care of the creation and deletion of processes, scheduling of various system resources to the different processes requesting them, and providing mechanisms for synchronization and communication among processes. 2. Memory Management. To execute a program, it must be loaded, together with the data it accesses, in the main memory (at least partially). To improve CPU utilization and to provide better responses time to its users, a computer system normally keeps several programs in main memory. The memory management module of an operating system takes care of the allocation and deal location of memory space to the various programs in need of this resource. 3. File Management. All computer systems are used for storage, retrieval and sharing of information. A computer normally stores such information in units called files. Processes can read information from files and can create new files for storing newly generated information. Information stored ii files is made persistent by storing them on a secondary storage media such as a magnetic disk. Files provide a natural and easy means of information sharing. That is, a file can be created by on application-and then shared with different applications at a later time. The file management module of an operating system takes care of filerelated activities such as organization, storing, retrieval, naming, sharing, and protection of files. It allows programs to use a set of operations that characterize the file abstraction and free the programmers from concerns about the details of space allocation and layout of the secondary storage device. 4. Device Management. A computer system normally consists of several I/O devices such as terminal, printer, disk, and tape. The device management module of an operating system takes care of controlling all the computer's I/O devices. It keeps track of I/O requests from processes, issue commands to the I/O devices, and ensures correct data transmission to/from an I/O device. It also provides an interface between the devices and the rest of the system that is simple and easy to use. Often, this interface is device independent, that is, the interface is same for all types of I/O devices. 5. Security. Computer systems often store large amounts of information, some of which is highly sensitive and valuable to their users. Users can trust the system and rely on it only if the various resources and information of a computer system are protected against destruction and unauthorized access. The security module of an operating system ensures this. This module also ensures that when several disjoint processes are being executed simultaneously, one process does not interfere with the others, or with the operating system itself. 6. Command Interpretation. A user communicates with the operating system, for using the various system resources, via a set of commands provided by the
operating system. The operating system also provides a simple language, known as command language (CL) or job control language (JCL using which a user can put several commands together from the command set to describe the resource requirements of the job. The command interpretation module of an operating system takes care of interpreting user commands, supplied individually or in the form of command language, and directing the system resources to handle the requests. With this mode of interaction with the system, the user is usually not too concerned with the hardware details of the system, or with how the operating system will direct the hardware to handle certain requests. In addition to the above listed major functions, an operating system also performs few other functions such as keeping an account of which users (or processes) use how much and what kinds of computer resources, maintenance of log of system usage by all users, and maintenance of internal time clock.
MEASURING SYSTEM PERFORMANCE The efficiency of an operating system and the overall performance of a computer system is usually measured in terms of the following: 1. Throughput. Throughput is the amount of work that the system is able to do per unit time. It is measured as the number of processes that are completed by the system per unit time. For example, if n processes are completed in an interval of t seconds, the throughput is taken as n/t processes per second during that interval. Throughput is normally measured in processes/hour. Note that the value of throughput does not depend only on the capability of a system, but also on the nature of jobs being processed by the system. For long processes, throughput may be one process/hour; and for short processes, throughput may be 100 processes/hour. 2. Turnaround time. From the point of view of an individual user, an important criterion is how long it takes the system to complete a job submitted by him/her. Turnaround time is the interval from the time of submission of a job to the system for processing to the time of completion of the job. Although higher throughput is desirable from the point of view of overall system performance, individual users are more interested in better turnaround time for their jobs. 3. Response time. Turnaround time is usually not a suitable measure for interactive systems, because in an interactive system a process can produce some output fairly early during its execution and can continue executing while previous results are being output to the user. Thus another measure used in case of interactive systems is response time, which is the interval from the time of
submission of a job to the system for processing to the time the first response for the job is produced by the system. In any computer system, it is desirable to maximize throughput and to minimize turnaround time and response time. PROCESS MANAGEMENT A process (also called job) is a program in execution. The main objective of the process management module of an operating system is to manage the processes submitted to the system in such a manner so as to minimize the idle time of the various processors (CPU, I/O processors, etc.) of the computer system. In this section, we will learn about some of the mechanisms commonly used in modern operating systems to achieve this objective. We will also see how these mechanisms have gradually evolved from the early days of computers. Process Management in Early Systems In early computer systems, a job was typically executed in the following manner: 1. A programmer would first write the program on paper. 2. It was then punched on cards or paper tape along with its data. 3. The deck.of cards or the paper tape containing the program and data was then submitted at tin reception counter of the computer centre. 4. An operator would then take the card deck or paper tape and manually load it into the system from card reader or paper tape reader. The operator was also responsible for loading any other software resource (such as a language compiler) or setting hardware devices required for the execution of the job. Before loading of the job, the operator had to use the front panel switches of the compute system to clear the main memory to remove any data remaining from the previous job. 5. The operator would then set the appropriate switches in the front panel to run the job. 6. The result of execution of the job was then printed on the printer, which was brought by the operator to the reception counter, so that the programmer could collect it later. The same process had to be repeated for each and every job to be executed by the computer. This method of job execution was known as the manual loading mechanism because the jobs had to be manually loaded one after another by the computer operator in the computer system. Notice that in this method, job-to-job transition was n automatic. The manual transition from one job to another caused lot of computer time to be wasted since the computer remained idle while the operator loaded and unloaded jobs and prepared the system for a new job. In order to reduce this idle time of the computer, a method of automatic job-to-job transition was devised. In this method, known as batch processing, when one job
is finished, the system control is automatically transferred back to the operating system which automatically performs the housekeeping jobs (such as clearing the memory remove any data remaining from the previous job) needed to load and run the next job. In case of bat processing systems, jobs were typically executed in the following manner: 1. Programmers would prepare their programs and data on card decks or paper tapes and submitted them at the reception counter of the computer centre. 2. The operator could periodically collect all the submitted programs and would batch them together and then load them all into the input device of the system at one time. 3. The operator would then give a command to the system to start executing the jobs. 4. The jobs were then automatically loaded from the input device and executed by the system one-by-one without any operator intervention. That is, the system would read the first job from the input device, execute it, print out its result on the printer, and then repeat these steps for each subsequent job till all the jobs in the submitted batch of jobs were over. 5. When all the jobs in the submitted batch were processed, the operator would separate the printed output for each job and keep them at the reception counter so that the programmers could collect them later. The batch processing mechanism helped in reducing the idle time of a computer system to a great extent because transition from one job to another did not require any operator intervention. Another major time saving made possible by batch processing was reduction in set-up time by batching of similar jobs together by the operator. For example, if all FORTRAN compilation jobs are batched together, the system needs to load the FORTRAN compiler only once for processing all these jobs. The obvious question is how the computer separates one job from another from a batch of jobs for automatic job-to-job transition. Moreover, how does the system know which compiler or what hardware devices are to be used by a particular job when there is no operator intervention? To facilitate all these, control statements and job control languages (JCLs) were introduced along with the concept of batch processing. The control statements are used by the operating system to identify a new job and to determine what resources are needed by the job during its execution. These control statements are written in a language known as the job control language. Usually every program and data sets are preceded and followed by JCL statements. When a program is prepared for a computer run, it is necessary to prepare the job control statements and place them in proper order along with the program, before the program is fed to the computer system. Thus, each program has, besides the
program itself, a set of instructions called JCL instructions, which instruct the operating system on the identity and requirements of the job. JCL statements tell the operating system things such as the name of the job, the user's name and account number, the I/O devices to be used during processing, the assembler or compiler to be used if language translation is required, and so on. Figure 14.2 shows a sample deck of cards prepared for the compilation and execution of a COBOL program in a batch processing system. Notice from the figure that in order to distinguish a control card from data or program cards, they are identified by a special character or pattern on the card. Several systems used the dollar-sign character ($) in the first column to identify a control card, while IBM's JCL used slash marks (//) in the first two columns. Other systems used some other characters or codes. The basic idea in selection of these special characters or codes was that no program or data card should have these characters or code. SEND Data for program $RUN $LOAD COBOL program $COBOL $JOB, ONGC05839, USER=SINHA Multiprogramming In case of both manual loading and batch processing of jobs, the jobs are loaded into the system and processed one at a time. That is, once loaded, a job will remain in the main memory until its execution is completed and the next job will be loaded only after the completion of the previous job. As shown in Figure 14.3, in such a situation, the job which is currently loaded and is being executed will be the sole occupant of the user's area of the main memory (the operating system always resides in a part of the main memory) and it will have the CPU exclusively available for itself. The situation shown in Figure 14.3 is that of a uniprogramming system in which only one job is processed by the system at a time and all the system resources are exclusively available for the job until it completes. It was observed that a job does not need the CPU for the entire duration of its processing. This is because in addition to doing computation (for which CPU is needed), a job often needs to perform I/O operations (such as reading or writing some data to a tape or disk, waiting for some data to be input from the keyboard, and printing some results) during the course of its processing. In fact, depending on the CPU utilization during the course of processing, jobs are broadly classified into the following two types:
1. CPU-bound jobs. These jobs mostly perform numerical calculations, with little I/O operations. They are so called because they heavily utilize the CPU during the course of their processing. Programs used for scientific and engineering computations usually fall in this category of jobs. 2. I/O-bound jobs. These jobs normally input vast amount of data, perform very little computation, and output large amount of information. They are so called because during the course of their processing, their CPU utilization is very low and most of the time they perform I/O operations. Programs used for commercial data processing applications usually fall in this category of jobs. In a uniprogramming system, the CPU will be idle whenever the job that is currently being processed by the system performs I/O operations. With CPUbound jobs, the CPU idle time may not be significant, but for I/O-bound jobs the CPU may be idle 80 to 90% of the time. Moreover, since I/O devices are slower than the CPU by 20 to 100 times, the CPU idle time is significant even for CPUbound jobs that perform little I/O. The concept of multiprogramming was introduced to minimize the idle time of the CPU by organizing multiple jobs in the system so that the CPU always has something to execute. How this is done is explained below. Multiprogramming is the name given to the interleaved execution of two or more different and independent programs by the same computer. In Figure 14.3 we have been introduced to the notion of having two programs in the main memory at the same time - the operating system for overall system control and the user program for performing user's job. In multiprogramming, this concept is carried one step further by placing two or more user programs in the main memory and executing them concurrently. With multiple user programs simultaneously residing in the main memory, when a user program that was executing (using the CPU) starts performing I/O operations, the CPU is allocated to another user program in the main memory that is ready to use the CPU instead of allowing the CPU to be idle. The CPU switches from one program to another almost instantaneously. Thus in multiprogramming several user programs share the time of the CPU to kept it busy. It is important to note here that multiprogramming is not defined to be the execution of instructions from several programs simultaneously. Rather, it does mean that there are a number of programs available to the CPU (stored in the main memory) and that a portion of one is executed, then a segment of another, and so on. Although two or more user programs reside in the main memory simultaneously, the CPU is capable of executing only one instruction at a time. Hence at any given time, only one of the programs has control of the CPU and is
executing instructions. Simultaneous execution of more than one program with a single CPU is impossible. In some multiprogramming systems, only a fixed number of jobs can be processed concurrently (multiprogramming with fixed tasks) (MFT), while in others the number of jobs can vary (multiprogramming with variable tasks) (MVT). A typical scenario of jobs in a multiprogramming system is shown in Figure 14.4. At the particular time instance shown in the figure, job A is not utilizing the CPU since it is busy writing output data on to the disk (I/O operations). Hence the CPU is being utilized to execute job B, which is also present in the main memory. Job C, also residing in the main memory, is waiting for the CPU to become free. Actually, as shown in Figure 14.5, in case of multiprogramming all the jobs residing in the main memory will be in one of the following three states -running (it is using the CPU), blocked (it is performing I/O operations) and ready (it is waiting for CPU to be assigned to it). In our example, jobs A, B and C are in blocked, running and ready states respectively. Since job C is in the ready state, as soon as the execution of job B is completed or job B requires doing I/O operation, the CPU will start executing job C. In the meanwhile, if job A completes its output operation, it will be in the ready state waiting for the CPU. Thus in multiprogramming, the CPU will never be idle as long as there is always some job to execute. Note that although many jobs may be in ready and blocked states, only one job can be running at any instant. The area occupied by each job residing simultaneously in the main memory is known as a memory partition. The actual number of partitions, and hence jobs, allowed in the main memory at any given time varies depending upon the operating system in use. Moreover, those jobs awaiting entry into main memory are queued on a fast secondary storage device such as a magnetic disk. The first job from this queue will be loaded into the main memory as soon as any one of the jobs already occupying the main memory is completed and the corresponding memory partition becomes free. Requirements of Multiprogramming Systems Multiprogramming systems have better throughput than uniprogramming systems because the CPU idle time is drastically reduced. However, multiprogramming systems are fairly sophisticated because they require the following additional hardware and software features: 1. Large memory. For multiprogramming to work satisfactorily, large main memory is required to accommodate a good number of user programs along with the operating system. 2. Memory protection. Computers designed for multiprogramming must provide some type of memory protection mechanism to prevent a job in one memory
partition from changing information or instruction of a job in another memory partition. For example, in Figure 14.4 we would not want job A to inadvertently destroy something in the completely independent job B or job C. In a multiprogramming system this is achieved by the memory protection feature, a combination of hardware and software, which prevents one job from addressing beyond the limits to its own allocated memory area. 3. Job status preservation. In multiprogramming, when a running job gets blocked for I/O processing the CPU is taken away from this job and is given to another job that is ready for execution. At a later time the former job will be allocated the CPU to continue its execution. Notice that this requires preserving of the job's complete status information when the CPU is taken away from it and restoring this information back before the CPU is given back to it again. To enable this, the operating system maintains a process control block (PCB) for each loaded process. A typical process control block is shown in Figure 14.6. With this arrangement, before taking away the CPU from a running process, its status is preserved in its PCB, and before the process resumes execution when the CPU is given back to it at a later time, its status is restored back from its PCB. Thus the process can continue execution without any problem. process identifier process state program counter values of various CPU registers accounting and scheduling information I/O status information 4. Proper job mix. A proper mix of I/O-bound and CPU-bound jobs is required to effectively overlap the operations of the CPU and I/O devices. If all the loaded jobs need I/O at the same time, the CPU will again be idle. Hence the main memory should contain some CPU-bound and some I/O-bound jobs so that at least one job is always ready to utilize the CPU. 5. CPU scheduling. In a multiprogramming system, often there will be situations in which two or more jobs will be in the ready state waiting for CPU to be allocated for execution. When more than one process is in the ready state when the CPU becomes free, the operating system must decide which of the ready jobs should be allocated the CPU for execution. The part of the operating system concerned with this decision is called the CPU scheduler, and the algorithm it uses is called the CPU scheduling algorithm. Multitasking
Technically speaking, multitasking, is the same as multiprogramming. That is, multitasking is the system's capability to concurrently work on more than one task (job or process). This means that whenever a task (job or process) needs to perform I/O operations, the CPU can be used for executing some other task (job or process) that is also residing in the system and is ready to use the CPU. Many authors do not distinguish between multiprogramming and multitasking because both the terms refer to the same concept. However, some authors prefer to use the term multiprogramming for multi-user systems (systems that are simultaneously used by many users such as mainframe and server class systems), and multitasking for single-user systems (systems that are used by only one user at a time such as a personal computer or a notebook computer). Note that even in a single-user system, it is not necessary that the system works Only on one job at a time. In fact, a user of a single-user system often has multiple tasks concurrently processed by the system. For example, while editing a file in the foreground, a sorting job can be given in the background. Similarly, while compilation of a program is in progress in the background, the user may be reading his/her electronic mails in the foreground. In this manner, a user may concurrently work on many tasks. In such a situation, the status of each of the tasks is normally viewed on the computer's screen by partitioning the screen into a number of windows. The progress of different tasks can be viewed on different windows in a multitasking system. Hence for those who like to differentiate between multiprogramming and multitasking, multiprogramming is the concurrent execution of multiple jobs (of same or different users) in a multi-user system, while multitasking is the concurrent execution of multiple jobs (often referred to as tasks of same user) in a single-user system. Multithreading Threads are a popular way to improve application performance. In traditional operating systems, the basic unit of CPU utilization is a process. Each process has its own program counter, its own register states, its own stack, and its own address space (memory, area allocated to it). On the other hand, in operating systems, with threads facility, the basic unit of CPU utilization is a thread. In these operating systems, a process consists of an address space and one or more threads of control [see Figure 14.7(b)]. Each thread of a process has its own program counter, its own register states, and its own stack. But all the threads of a process share the same address space. Hence they also share the same global variables. In addition, all threads of a process also share the same set of operating system resources, such as open files, signals, accounting information, and so on. Due to the sharing of address space, there is no protection between the threads of a process. However,
this is not a problem. Protection between processes is needed because different processes may belong to different users. But a process (and nence all its threads) is always owned by a single user. Therefore, protection between multiple threads of a process is not necessary. If protection is required between two threads of a process, it is preferable to put them in different processes, instead of putting them in a single process. Threads share a CPU in the same way as processes do. At a particular instance of time, a thread can be in any one of several states - running, blocked, ready, or terminated. Due to these similarities, threads are often viewed as miniprocesses. In fact, in operating systems with threads facility, a process having a single thread corresponds to a process of a traditional operating system [see Figure 14.7(a)]. Threads are often referred to as lightweight processes and traditional processes are referred to as heavyweight processes. Motivations for Using Threads The main motivations for using a multithreaded process instead of multiple singlethreaded processes for performing some computation activities are as follows: 1. The overheads involved in creating a new process are, in general, considerably greater than those of creating a new thread within a process. This is mainly because when a new process is created, its address space has to be created from scratch, although a part of it might be inherited from the process's parent process. However, when a new thread is created, it uses the address space of its process that need not be created from scratch. 2. Due to the sharing of address space and other operating system resources among the threads of a process, the overhead involved in CPU switching among peer threads is very small as compared to CPU switching among processes having their own address spaces. This is the reason why threads are called lightweight processes. 3. Resource sharing can be achieved more efficiently and naturally between threads of a process than between processes because all threads of a process share the same address space. 4. Finally, the use of threads is also motivated by the fact that a set of threads using a shared address space is the most natural way to program many applications. For example, in an application that uses the producer-consumer model, the producer and the consumer processes must share a common buffer. Therefore, programming the application in such a way that the producer and consumer are two threads of the same process makes the software design simpler. Multiprocessing
Up to this point we have considered systems with a single CPU. However, we have already seen that the use of I/O processors improves the efficiency of a computer system by making possible concurrent input, processing, and output operations. The CPU can perform arithmetic and logical operations on parts of one or more programs while I/O operations are concurrently carried out by I/O processors on other parts of programs. Figure 14.8 shows the architecture of a computer with its CPU, memory and I/O processors. The idea of use of I/O processors to improve the performance of a computer system was carried one step further by designing systems that make use of more than one CPU. Such systems are called multiprocessing systems. The term multiprocessing is used to describe interconnected computer configurations or computers with two or more CPUs that have the ability to simultaneously execute several programs. In such a system, instructions from different and independent programs can be processed simultaneously by different CPUs or the CPUs may simultaneously execute different instructions from the same program. The basic organization of a typical multiprocessing system is shown in Figure 14.9. Tightly and Loosely Coupled Multiprocessing Systems Multiprocessing systems are basically of two types - tightly-coupled systems and loosely-coupled systems. In tightly-coupled systems, there is a single system-wide primary memory that is shared by all the processors. On the other hand, in looselycoupled systems, the processors do not share memory, and each processor has its own local memory. In contrast to the tightly-coupled systems, the processors of loosely-coupled systems can be located far from each other to cover a wider geographical area. Difference between Multiprogramming and Multiprocessing Multiprogramming is the interleaved execution of two or more processes by a single-CPU computer system. On the other hand, multiprocessing is the simultaneous execution of two or more processes by a computer system having more than one CPU. To be more specific, multiprogramming involves executing a portion of one program, then a segment of another, etc., in brief consecutive time periods. Multiprocessing, however, makes it possible for the system to simultaneously work on several program segments of one or more programs. Advantages and Limitations of Multiprocessing Multiprocessing systems typically have the following advantages:
1. Better Performance. Due to multiplicity of processors, multiprocessor systems have better performance than single-processor systems. That is, the multiple processors of such a system can be utilized properly for providing shorter response times and higher throughput than a single-processor system. For example, if there are two different programs to be run, two processors are evidently more powerful than one because the programs can be simultaneously run on different processors. Furthermore, if a particular computation can be partitioned into a number of subcomputations that can run concurrently, in a multiprocessor system all the subcomputations can be simultaneously run with each one on a different processor. However, the speed-up ratio with n processors is not n, but less than n. This is because when multiple processors cooperate to execute the programs or subprograms, a certain amount of overhead is incurred in keeping everything working correctly. This overhead, plus contention for shared resources, lowers the expected gains from additional processors. 2. Better Reliability. Due to multiplicity of processors, multiprocessor systems also have better reliability than single-processor systems. In a properly designed multiprocessor system, if one of the processors breaks down, the other processor(s) automatically takes over the system workload until repairs are made. Thus a complete breakdown of such systems can be avoided. For example, if a system has 4 processors and one fails, then the remaining 3 processors can be utilized to process the jobs submitted to the system. Thus, the entire system runs only 25% slower, rather than failing altogether. This ability of a system to continue providing service proportional to the level of non-failed hardware is called graceful degradation feature. Multiprocessing systems, however, require a very sophisticated operating system to schedule, balance, and coordinate the input, output, and processing activities of multiple processors. The design of such an operating system is a complex and time taking job. Moreover, multiprocessing systems are expensive to procure and maintain. In addition to the high charge paid initially, the regular operation and maintenance of these systems is also a costly affair. Time-sharing Time-sharing is a mechanism to provide simultaneous interactive use of a computer system by many users in such a way that each user is given the impression that he/she has his/her own computer. It uses multiprogramming with a special CPU scheduling algorithm to achieve this. A time-sharing system has many (even hundreds of) user terminals simultaneously connected to the same computer. Using these terminals, multiple users can
simultaneously work on the system. The multiprogramming feature allows multiple user programs to simultaneously reside in the main memory. The special CPU scheduling algorithm used in a time-sharing system allocates a very short period of CPU time one-by-one to each user process, beginning from the first user process and proceeding through the last one, and then again beginning from the first one. This short period of time during which a user process gets the attention of the CPU is known as a time slice, time slot, or quantum and is typically of the order of 10 to 100 milliseconds. That is, when the CPU is allocated to a user process, the user process will use the CPU until the allotted time slice expires (the system's clock sends an interrupt signal to the CPU after every time slice), or until the process needs to perform some I/O operation, or if the execution of the process is over during this time period. Notice here that in a time-sharing system, the CPU is taken away from a running process when the allotted time slice expires, even though the process can continue to run. Hence the process state diagram of a timesharing system is as shown in Figure 14.10 (compare this with the process state diagram of Figure 14.5). Now let us see how the CPU scheduling algorithm mentioned above gives an impression to each user that he/she has his/her own computer. Let us assume that the time slice of a time-sharing system is 10 milliseconds. That is the CPU scheduling algorithm of this system allocates 10 milliseconds to each user process one-by-one in a circular fashion (when the last process is over, it comes back to the first process). Suppose the processing speed of the system's CPU is of the order of 500 million instructions per second. That is, it can execute 500 x 106 x 103 x 10 = 5x106 = 5 million instructions in 10 milliseconds. This is large enough for substantial progress of a single user process. Now suppose there are 100 user terminals in this time-sharing system and 100 users are simultaneously using the system. Then if 10 milliseconds is allocated to each user process one-by-one, a particular user will get the CPU's attention once in every 10 x 100 milliseconds = 1 second. As human reaction time is normally of the order of a few seconds, a particular user will not notice any delay in executing his/her commands and will normally feel that he/she is the sole user of the system. In this manner, each user is given the impression that he/she has his/her own computer, whereas actually a single computer is shared among many uses. Requirements of Time-sharing Systems Time-sharing systems typically require the following additional hardware and software features: 1. A number of terminals simultaneously connected to the system so that multiple users can simultaneously use the system in interactive mode. 2. A relatively large memory to support multiprogramming.
3. Memory protection mechanism to prevent one job's instructions and data from other jobs in a multiprogramming environment. 4. Job status preservation mechanism to preserve a job's complete status information when the CPU taken away from it and restoring this information back before the CPU is given back to it again. 5. A special CPU scheduling algorithm that allocates a very short period of CPU time one-by-one to each user process in a circular fashion. 6. An alarm clock mechanism to send an interrupt signal to the CPU after every time slice. Advantages of Time-sharing Systems Although time-sharing systems are complex to design, they provide several advantages to their users. The main advantages of time-sharing system are as follows: 1. Reduces CPU idle time. It will be waste of precious CPU time to allow a single user to use a large computer interactively because his/her speed of thinking and typing is much slower than the processing speed of a computer. While a particular user is engaged in thinking or typing his/her input, a time-sharing system can service many other users. In this manner, time-sharing systems help in reducing the CPU idle time to a great extent, increasing the system throughput. 2. Provides advantages of quick response time. The special CPU scheduling algorithm used in time-sharing systems ensures quick response time to all users. This feature allows users to interact with the system more rapidly while working on their problem. For example, a time-sharing system can be effectively used for interactive programming and debugging to improve programmers efficiency. Multiple programmers can simultaneously proceed step-by-step, writing, testing and debugging portions of their programs or trying out various approaches to a problem solution. The great benefit of such a system is that errors can be encountered, corrected, and work can continue immediately for all the simultaneous users of the system. This is in contrast to a batch system in which errors are corrected offline and the job is resubmitted for another run. The time delay between job submission and return of the output in a batch system is often measured in hours. 3. Offers good computing facility to small users. Small users can gain direct access to much more sophisticated hardware and software than they could otherwise justify or afford. In time-sharing systems, they merely pay a fee for resources used and are relieved of the hardware, software, and personnel problems associated with acquiring and maintaining their own installation. MEMORY MANAGEMENT
Other than the CPU, the main memory is an important resource of a computer system that must be properly managed for the overall system performance. The memory management module of an operating system takes care of this requirement. Its job is to keep track of which parts of memory are in use and which parts are not in use, to allocate memory to processes when they need it and deallocate when they are done. In this section we will discuss about a number of different memory management schemes used in earlier and modern operating systems. Uniprogramming Memory Model The uniprogramming memory model is used in uniprogramming systems in which only one job is processed by the system at a time and all the system resources are exclusively available for the job till it completes. As shown in Figure 14.11, in this memory management scheme, the operating system resides in one part of the memory and the entire remaining part of the memory is available for use by the currently active user process. In this scheme, the operating system loads a program to be executed from disk into the user area of the memory and executes it. When the process finishes, the operating system cleans up the user area of the memory and then loads the next program to be executed. Although this memory management scheme is simple and easy to implement, it does not lead to proper utilization of the available main memory resource. This is because whatever memory space is not occupied by the currently active user process remains unusued for the entire duration of the process's execution. Hence this memory management scheme is now used only on very small or dedicated computer systems. Multiprogramming Memory Models We saw that in a multiprogramming system multiple user processes need to simultaneously reside in the main memory. The two memory management schemes used to facilitate this are multiprogramming with fixed number of memory partitions and multiprogramming with variable number of memory partitions. They are described below. Multiprogramming with Fixed Number of Memory Partitions In this scheme, the user area of the memory is divided into a number of fixed-sized partitions. The partitions may be of equal or unequal size, but the size, of each partition is fixed. Figure 14.12 shows a multiprogramming memory model with n equal-sized partitions. Each partition may contain exactly one process. Thus, the
degree of multiprogramming is bounded by the number of partitions. That is, at a time only n processes can be loaded in the system. All new jobs are put into an input queue. When a partition is free, the next job from the input queue is picked up and loaded into the free partition. When a process terminates, the partition occupied by it becomes free for use by another process waiting in the input queue. Note that if the system uses unequal-sized partitions, when a partition becomes free, the process closest to the front of the input queue that fits in it could be loaded into this partition. This scheme of memory management was used in IBM OS/360 mainframe systems for several years. It was called MFI (Multiprogramming with a Fixed number of Tasks). It is no longer in use. Multiprogramming with Variable Number of Memory Partitions In the scheme described above, since the partitions are of fixed size, any space in a partition that is in excess of actual memory requirement of the process loaded into it is lost (remains unused) till the process terminates. On an average, 50% of the memory may remain unused due to this reason. To overcome this problem of under utilization of memory, the scheme with variable number of memory partitions was introduced. In this the scheme, number, size, and location of the partitions vary dynamically as processes come and go. This memory management scheme is illustrated in Figure 14.13 with an example. Initially, all memory in the user area is available for user processes. When a process arrives, only as much memory as is needed by it is allocated to it, keeping the rest available to satisfy future requests. As processes come and go, memory partitions get allocated and deallocated, each partition being exactly of the size of the memory required by the process to which it is allocated. Since the memory requirement of different processes is generally different, as processes come and go, various sizes of free memory blocks get created throughout the memory. The operating system maintains a table to keep track of which parts of memory are free and which are occupied. When a new process arrives, the operating system searches for a free block that is large enough for this process. If the free block is too large, it is split into two parts. One part is large enough to meet the memory requirement of the process and is allocated to the process. The other part contains the remaining memory that is entered as a smaller free block in the operating system table. When a process terminates, it releases the partition allocated to it. If the released partition is adjacent to a free block or blocks, it is merged with the free block or blocks to create a larger free block. In this manner, the number, size, and location of the partitions vary dynamically as processes come and go. Memory Fragmentation
External Fragmentation Notice from Figure 14.13 that as processes come and go in a system that uses the scheme of variable number of memory partitions, the free memory blocks get broken into smaller pieces scattered throughout the memory. External fragmentation is a situation when enough total free memory space exists to satisfy the memory need of a process, but still the process cannot be loaded because the available free memory is not contiguous (is fragmented into multiple free blocks of memory). For example, in case of Figure 14.13(h), if the sizes of free blocks Free 1, Free 2 and Free 3 are 400K, 100K and 100K respectively, then the total available free memory is 600K. Now if a new process arrives whose memory requirement is 500K, it cannot be loaded because 500K of contiguous free memory space is not available. Thus this is an example of external fragmentation. The amount of unusable memory area due to external fragmentation depends on the total size of main memory and the average memory requirement of the processes. However, statistical analysis indicates that as much as one-third of memory may be unusable due to external fragmentation. Internal Fragmentation Internal fragmentation is a situation when a process is allocated more memory than its actual memory requirement and the additional memory allocated to the process remains unutilized because it is neither used by the process to which it is allocated nor it can be allocated to any other process for use. Internal fragmentation may occur in the following situations: 1. When the system uses fixed number of fixed-sized memory partitions, any space in a partition that is in excess of the actual memory requirement of the process loaded into it is an internally fragmented memory space. Figure 14.14 illustrates this situation with an example. Note that, on an average, 50% of the memory may remain unused in such systems due to internal fragmentation. 2. When the system uses variable number of variable-sized memory partitions, it may happen that there is a free block of size 200K and a new process arrives whose memory requirement is 199.5K. If we allocate exactly the requested size of memory, we will be left with a free block of size 0.5K. Considering the overhead involved in keeping track of this 0.5K free block, the general approach is to allocate such small memory areas as part of the process's memory request. Thus, the allocated memory may be slightly larger than the requested memory, and the additional memory that gets allocated to a process in this way is an internally fragmented memory space. Paging
We saw in the discussion above that memory compaction is a mechanism used to solve the problem of external fragmentation. We also saw that memory compaction is an expensive operation requiring lot of CPU time because it involves copying memory contents from one memory location to another to make scattered free space contiguous. Paging is another mechanism used to solve the problem of external fragmentation. It is considered to be better than the memory compaction mechanism because it solves the problem of external fragmentation without the need to make scattered free space contiguous. Rather it allows a process's memory to be noncontiguous, thus allowing a process to be allocated physical memory wherever free memory blocks are available. Because of its advantage over memory compaction mechanism, paging mechanism (in its various forms) is used in many modern operating systems. How this mechanism works is explained below. In this mechanism, the physical memory of a computer system is partitioned into fixed-sized blocks called page frames. The total memory requirement of a process (including its instructions and data) is also partitioned into blocks of the same size called pages. Page sizes are typically of the order of 1 Kilobytes to 8 Kilobytes and vary from one system to another. Now when a process is to be loaded into the memory, its pages are loaded into the free page frames wherever free page frames are available and a mapping table (called page table) is maintained by the operating system to keep track of which page of the process is loaded into which page frame of the memory. Atypical example to illustrate this is shown in Figure 14.15. Notice from the figure that the page frames allocated to the pages of the process need not be contiguous. To ensure correct memory addressing, in case of paging every address generated by the CPU is divided into two parts consisting of a page number and a page offset. The page table is used to obtain the corresponding page frame number from the page number, and the page offset is added to the base address of the page frame number to define the physical memory address that is sent to the memory unit. Notice that there is no external fragmentation problem in a system that uses the paging scheme because free pa frames allocated to a process can be anywhere in the memory (need not be contiguous). However, the paging scheme leads to some unutilized memory space due to internal fragmentation because the unit of memory allocation is a page frame. If the memory requirements of a process do not happen to fall on page boundaries, the last page frame allocated to the process will not be completely full. For example, if the page size used is 2048 bytes, and a process's memory requirement is 23,548 bytes, the process will need 11 pages plus 1020 bytes. As
the unit of memory allocation is a page frame, the process will be allocated 12 page frames, resulting in an internal fragmentation of 2048 - 1020 = 1028 bytes. On an average, internal fragmentation leads to wastage of one-half page per process in case of paging scheme. Hence, the larger is the page size, the larger will be the memory wastage due to internal fragmentation. This suggests the use of small page sizes. However, the smaller the page size, the larger will be the page tables and the overhead involved in maintaining them. Hence these two factors have to be properly balanced to decide a suitable page size. Virtual Memory What is Virtual Memory? In all the memory management schemes discussed till now, a process was loaded into the main memory for execution only when sufficient free memory (contiguous or non-contiguous) was found to be available to load the entire process. These schemes suffer from the following two main limitations: 1. A process cannot be loaded and has to keep waiting for its execution to start until sufficient free memory for loading the entire process becomes available. This may delay a process's turnaround time to a great extent. 2. A process cannot be loaded (and hence executed) in a system whose main memory size is less than the total memory required by the process. Virtual memory is a memory management scheme that overcomes the above mentioned limitations by allowing the execution of processes that might not be completely loaded in the main memory. That is, it does not require the entire process to be in memory before the process can execute. How is Virtual Memory Realized? The three basic concepts used for the realization of virtual memory are: 1. On-line secondary storage. It is a secondary storage device whose capacity is much larger than the main memory capacity and which is always kept on-line to the system. A high-speed disk storage is usually used for this purpose. 2. Swapping. Swapping is the process of transferring a block of data from the on-line secondary storage to main memory or from main memory to the on-line secondary storage. When data is transferred from on-line secondary storage to main memory, it is called swapping in of data, and when data is transferred from main memory to on-line secondary storage, it is called swapping out of data. 3. Demand paging. Demand paging is a scheme similar to that of paging scheme with an additional facility of swapping in of pages of a process as and when they are needed during the execution of the process, rather than loading all the pages before starting the process's execution. When there is no free page frame in the
memory to accommodate a page that needs to be swapped in to continue the process's execution, a page-replacement algorithm is invoked to create one for the accessed page. It deals with the process of selecting a victim page that is residing in the memory but is not currently being used. The selected victim page is swapped out to free the page frame it is occupying, which is then used to hold the accessed page of the process. Hence in a virtual memory system, all processes are partitioned into pages and reside on the on-line secondary storage. The physical memory is also partitioned into page frames of the same size. Now instead of swapping in the entire process before its execution can start, a swapping algorithm (called demand paging) is used that swaps in only those pages of a process that are currently needed in the memory for continuing the process's execution. This idea is based on the observation that since the instructions of a program are executed one-by-one, all parts of the program are not needed simultaneously in the memory. Based on the way virtual memory is realized, virtual memory is also often described as a hierarchy of two storage systems - one of them is a low cost, large capacity, low speed system (on-line disk storage), and the other is a high cost, small capacity, high speed system (main memory). The operating system manages the two storage systems in such a way that the users of the system feel that they have access to a single, large, directly addressable and fast main memory. That is, from the point of view of applications programmers, the effective (or virtual) size of the available main memory appears to be unlimited. Advantages and Disadvantages of Virtual Memory Advantages Virtual memory typically has the following advantages: 1. It enables separation of logical memory and physical memory. With this mechanism, a very large virtual memory can be provided for programmers on a system having smaller physical memory. That is, the logical address space is no longer constrained by the physical memory size of the system. 2. It enables the execution of a process on a system whose main memory size is less than the total memory required by the process. 3. It enables a process's execution to be started even when sufficient free memory for loading the entire process is not available. This helps in greatly improving the process's turnaround time.
4. With virtual memory, a program would no longer be constrained by the amount of physical memory that is available. This makes programming task much easier, since the programmer no longer needs to worry about the size limitations of the physical memory available, but can concentrate instead on the problem to be programmed. 5. An analysis of the execution of real programs shows that programs often do not execute all parts of their code during a typical execution run. For example, depending on the data, a large part of the code may be skipped on some condition check and may never be executed during that execution run. Similarly, certain options and features of a program, such as those for handling unusual error conditions, may be rarely used, and are generally not needed during most execution runs of the program. With virtual memory mechanism, since the parts of a program are loaded on demand, the parts that are not needed during a particular execution run of a program may never get loaded. This in turn leads to less I/O activity than in the case when the whole program is loaded before execution. With less I/O, each user program would run faster leading to better throughput, turnaround time, and response time. 6. Virtual memory enables a process's execution to be started even when sufficient free memory for loading the entire process is not available. Due to this, virtual memory can be effectively used to simultaneously accommodate program segments of a large number of users in the main memory. This increases the degree of multiprogramming resulting in increased CPU utilization and system throughput. Disadvantages Virtual memory typically has the following disadvantages: 1. It is difficult to implement because it requires algorithms to support demand paging. 2. If used carelessly, it may substantially decrease performance instead of increasing performance. This happens when page-fault rate is very high for a process. That is, the process spends more time in swapping out and swapping in of pages than in its execution. FILE MANAGEMENT A file is a collection of related information. Every file has a name, its data, and attributes. The name of a file uniquely identifies it in the system and is used by its users to access it. A file's data is its contents. The contents of a file is a sequence of bits, bytes, lines, or records whose meaning is defined by the file's creator and user. The attributes of a file contain other information about the file such as the date and time of its creation, date and time of last access, date and time of last update, its current size, its protection features (who can access the file and in what
way), etc. The list of attributes maintained for a file varies considerably from one system to another. The file management module of an operating system takes care of file-related activities such as structuring, accessing, naming, sharing, and protection of files. In this section we will discuss about the important concepts used for file management in various operating systems. File Structure Files can be structured in many ways. However, the two popular types of file structures used by operating systems are files as unstructured sequence of bytes and files as sequence of fixed-length records. The two types of file structures are briefly described below. Files as Unstructured Sequence of Bytes As shown in Figure 14.16(a), in this case the operating system treats a file as nothing more than byte sequences. That is, all file access operations at operating system level deal with the number of bytes to be read, written or skipped. No interpretation of the bytes is made by the operating system. This type of file structuring scheme provides maximum flexibility, but minimal support. That is, each application program must include its own code to interpret an input file into the appropriate structure. This type of file structure is used in two of the most popular operating systems, UNIX and MS-DOS. Files as Sequence of Fixed-Length Records As shown in Figure 14.16(b), in this case the operating system treats a file as a sequence of fixed-length records. Each record has some internal structure defined by the creator of the file. In this case all file access operations at operating system level deal with the number of records to be read, written or skipped. This type of file structure was commonly used in olden days operating systems. For example, when the 80-column punched paper cards were popular as input media, and 132 column printers were popular as output device, many operating systems based their file systems on input files consisting of 80-character records and output files consisting of 132-charaeter records. The most recent operating system that used this type of file structuring scheme was CP/M. File Access Methods
To use the information stored in a file, it must be accessed and read into computer memory. The two commonly supported file access methods at operating system level are sequential and random access. They are briefly described below. 1. Sequential Access Files. Sequential access files are normally used with sequential access storage media such as magnetic tape. Information stored in a sequential access file can be accessed only sequentially. That is, a process can read the bytes or records in the file in the order in which they are stored, starting at the beginning. Reading of bytes or records randomly or out of order is not possible. A sequential file can, however, be rewound and read as often as needed. 2. Random Access Files. Random access files are normally used with random access storage media such as magnetic or optical disks. Unlike a sequential access file, information stored in a random access file can be accessed randomly irrespective of the order in which the bytes or records are stored. For example, in a record-oriented file, a specific record can be directly accessed by specifying the value of its key field, rather than reading the records one-by-one from the beginning and comparing its key field value with the value of the key field of the desired record till the desired record is found. Similarly, in a byte-oriented file, it is possible to directly go to the nth byte of the file without the need to read all the (n-1) bytes before it. Random access files are essential for many applications. For example, in a railway reservation system, the information about all the tickets booked on a particular train may be stored in a single file. If a customer wants to cancel an already booked seat, the reservation program must be able to access the specific record for that seat without having to read the records for hundreds of other booked seats first, so that the booking service can be fast enough. All operating systems do not support both sequential and random access files. Some support only sequential access files, some support only random access files, while there are some which support both. Those, which support files of both types, normally require that a file be declared as sequential or random when it is created; such a file can be accessed only in a manner consistent with its declaration. Most modern operating system support only random access files. File Operations An operating system provides a set of operations to deal with files and their contents. A typical set of file operations provided by an operating system may be as follows (the actual set varies from one operating system to another): 1. Create. Is used to create a new file. 2. Delete. Is used to delete an existing file that is no longer needed. 3. Open. An already existing file must be opened before using it. This operation is used to open an existing file when a user wants to start using it. 4. Close. When a user has finished using a file, the file must be closed using this operation.
5. Read. Is used to read data stored in a file. 6. Write. Is used to write new data in a file. 7. Seek. This operation is used with random access files to first position the read/write pointer to a specific place in the file, so that data can be read from, or written to, that position. 8. Get attributes. Is used to access the attributes of a file. 9. Set attributes. Is used to change the user-settable attributes (such as, protection mode) of a file. 10. Rename. Is used to change the name of an existing file. 11. Copy. Is used to create a copy of a file, or to copy a file to an I/O device such as a printer or a display. File Naming When a file is created, it is given a name, which can be later used to access the file. The rules for naming files vary from one operating system to another. For instance, consider the following examples: 1. MS-DOS allows only up to 8 characters for a file name, Macintosh allows up to 31 characters, and Microsoft Windows allows up to 255 characters. With longer file names, users have greater flexibility in assigning more descriptive and meaningful names to their files. 2. Some operating systems allow only letters and numbers to be used in file names, whereas others also allow special characters (sometimes barring a few) to be used in file names. Hence names like letter3 Jrom Ram, reminder2_to_Sham are often valid file names. 3. Some operating systems (such as Unix) distinguish between upper case letters and lower case letters, whereas others (such as MS-DOS) do not. Thus the file names Sinha, sinha and SINHA will be treated as different and can be assigned to three different files in Unix, but they will be treated as same in MS-DOS and cannot be assigned to three different files. 4. Many operating systems support file names having two or more parts, with the different parts separated by a period. For example, MS-DOS supports two-part file names in which the first part can have 1 to 8 characters, and the optional second part can have 1 to 3 characters (such as letter.txt). In Unix, a file may have more than two parts, as inprog.cz, where .c and .z are used to indicate that it is a C language program file that has been compressed using the Ziv-Lempel compression algorithm. In such file names, the second and subsequent parts are called file extensions. File extensions usually indicate something about the file and are often used by applications to check for the intended type of file before operating on it. For example, a C language compiler may actually insist that the files it is to compile end in 4c, and it may refuse to compile them if they do not. Some applications automatically supply their own extensions, such as .doc (for a
Microsoft Word file) or .wk4 (for a Lotus 1-2-3 file). Some typical file extensions and their meanings are shown in Figure 14.17. File extension .bas .c .ftn .pas .obj .bin .lib .dat .hip .man .txt .bak .doc .wav .wk4 .xls .jpg .gif
Its meaning Basic source program file C source program file Fortran source program file Pascal source program file Object file (compiler output, not yet linked) Executable binary program file Library of .obj files used by the linker Data file Text file for HELP command Online manual page file General text file Backup file Microsoft word document file Microsoft windows sound file Lotus 1-2-3 spreadsheet file Microsoft Excel spreadsheet file JPEG graphics file GIF graphics file
Figure 14.17. Some typical file extensions and their meaning. Directories Directories are used by operating systems to map file names to their corresponding file attributes and file data, and also to provide greater flexibility to users in file naming. How directories help in achieving these objectives is briefly explained below. Directory Contents A directory has a number of entries, one per file. As shown in Figure 14.18, each entry contains a file name, the attributes of the file or pointer to the data structure containing the attributes, and pointer to the file data (disk address where the file data is stored). File
File attributes or pointer to data Pointer to file
Name structure attributes
containing
file data (Disk address)
Figure 14.18. Typical contents of a directory. There is one entry per file. When a new file is created, an entry for it is created in the directory, and when a file is deleted, its corresponding entry is deleted from the directory. When an existing file is opened for use, the operating system searches its corresponding entry in the directory and then extracts its attributes and disk addresses, and loads them in a table in main memory. All subsequent references to the file use this information in main memory. Directory Organization The number of directories used and the way these directories are organized varies from one system to another. The following three types of organizations are commonly used: 1. Single directory. This is the simplest organization. As shown in Figure 14.19(a), in this case, a single directory contains entries for all the files in the system. Although this organization is simple to implement, it has significant limitations when the number of files increases or when there are more than one users. As the number of files increases, it becomes difficult to remember the names of all the files in order to create new files with unique names. Similarly, in a system having multiple users, if two users choose the same file name, the system will not accept the file name of the user who did so later than the other user because it violates the unique-name rule. Such conflicts and confusions make this directory organization unsuitable for multi-user systems. Hence it is used only by the most primitive microcomputer operating systems. This organization is popularly known as flat directory structure. 2. One directory per user. As shown in Figure 14.19(b), in this organization, there is a separate directory for each user. The names of all the files belonging to a particular user are listed in his/her directory. As a file name needs to be unique only within a directory, different users can have files with the same name. This overcomes the limitation of file name conflicts among users in case of single directory organization. However, this organization is still not very satisfactory for users with many files because it becomes difficult for a user to remember the names of all the files created over a period of time when the number of files in the
user's directory becomes large. This organization is popularly known as two-level directory structure. 3. Multiple directories per user. As shown in Figure 14.19(c), in this organization also, there is a separate directory for each user. However, each user can create and use as many sub-directories or sub-sub-directories, or sub-sub-subdirectories, and so on to group his/her files in natural ways. For example, a user may like to group his/her files project-wise, in which case a sub-directory may be created for each project and all files pertaining to a particular project can be listed within the corresponding project's sub-directory. As a file name needs to be unique only within a directory, the same user can have files with the same name in different sub-directories. This feature provides greater flexibility to a user in naming his/her files because the files can be grouped together in logical ways and the user can remember the file name along with the group name to which it belongs even when the number of files for a user becomes large. This organization is popularly known as tree-structured hierarchical directory structure. Disk Space Allocation Methods Files are normally stored on disks for permanent storage. A disk normally contains many files. Disk space allocation methods deal with how to allocate space to multiple files on disk so as to meet the following two objectives: 1. Disk space is effectively utilized, and 2. Desired piece of information can be quickly accessed from files. The three commonly used methods for disk space allocation are contiguous, linked, and indexed. They are described below. However, before that, it is important to note that operating systems divide the total disk storage space into equal sized blocks (see Figure 14.20). The size of a block is normally a multiple of page size in systems that use paging mechanism for memory management. Disk space is allocated to files in units of blocks. It is possible to read a block from the disk, to modify the block, and write it back into the same place. Contiguous Allocation In this method, each file occupies a set of contiguous blocks on the disk. Thus if a disk has 2K blocks, a file of size 1 IK will be allocated 6 contiguous blocks. As shown in Figure 14.21, in this method, the disk addresses in the directory indicate the starting block number and the total number of blocks allocated to each file. Contiguous disk space allocation has the following advantages: 1. It is simple to implement. 2. Its performance is excellent because the entire file can be read from the disk in a single operation. 3. It easily supports both sequential and direct access because all the blocks allocated to a file contiguous.
File Name
File attributes
Disk address Starting Total number block of number blocks ! 0 3 19 6 29 8 5 4
A — B — C — D — Directory File 'A' Occupies 3 blocks numbered 0, 1 and 2. File 'B' occupies 6 blocks numbered 19, 20, 21, 22, 23 and 24. File 'C occupies 8 blocks numbered 29, 30, 31, 32, 33, 34, 35 and 36. File 'D' occupies 4 blocks numbered 5, 6, 7 and 8. Figure 14.21. Illustrating allocation of disk blocks to files in contiguous allocation method. This method, however, has the following drawbacks: 1. It suffers from external fragmentation problem. External fragmentation is a situation when enough total free disk blocks exist to satisfy the disk space requirement of a file, but still the file cannot be stored because the available free disk blocks are not contiguous. The amount of unusable disk space due to external fragmentation depends on the total disk space and the average disk space requirement of the files. However, statistical analysis indicates that as much as one-third of disk space may be unusable due to external fragmentation. To take care of this problem, systems which use contiguous disk space allocation method often provide the facility of disk compaction which moves files to bring all free blocks together. Disk compaction is a time consuming operation and may take several hour. Hence it is usually performed at night when the system is otherwise idle. 2. Another problem with this method is in determining how much space is needed for a file when it is created so that appropriate number of contiguous blocks can be found and allocated to the file. In general, it is difficult to estimate the size of an output file. If the size is over estimated, it will lead to unused disk space due to internal fragmentation. On the other hand, if the size is under estimated, the file will have to be moved completely to a new area on the disk when the file size grows beyond the originally estimated size. This often leads to external fragmentation problem and slower file access operations. Linked Allocation
In this method, the set of disk blocks allocated to a file are linked together in the form of a linked list of blocks as shown in Figure 14.22. The first word of each disk block is used as a pointer to the next one, and the rest of the words in the block are used to store file data. Note that the disk blocks allocated to a file may be scattered anywhere on the disk. As shown in Figure 14.22, in this method, the disk address in the directory indicates the number of the first block allocated to each file. Linked disk space allocation method has the following advantages: 1. There is no external fragmentation in this method. Every disk block can be used. Since there is no loss of disk space due to external fragmentation, there is no need to implement disk compaction facility. 2. There is no need to determine and declare the total anticipated size of a file at the time of its creation. A file can freely grow at any time since new disk blocks required to be allocated to the file may be anywhere on the disk. The method, however, has the following drawbacks: 1. The method can be used effectively only for sequential-access files. Random access of files in this method is extremely slow. For example, to access some data from the ih block of a file, we must start at the first block and follow the pointers until we get to the ith block. This is slow because each access to a pointer requires accessing a new block that requires a disk read operation. 2. Another problem is reliability. Since the blocks allocated to a file are linked together by pointers scattered all over the disk, if a pointer were lost due to disk corruption, all file data beyond that disk block would become inaccessible. This would lead to garbage blocks, blocks that are neither free nor being used, and would require implementation of a garbage collection algorithm. This algorithm searches for garbage blocks and adds them to the list of free disk blocks. Indexed Allocation In this method also, the disk blocks allocated to a file may be scattered anywhere on the disk. However, unlike linked allocation in which the set of blocks allocated to a file are linked together in a linked list, in this method the pointers to the set of blocks allocated to a file are stored together in a special block known as index block. Thus, each file has its own index block, which is an array of disk block addresses. The i'h entry in the index block points to the i'h block of the file. As shown in Figure 14.23, in this method, the disk address in the directory indicates the number of the index block allocated to each file. Indexed disk space allocation method has the following advantages: 1. There is no external fragmentation in this method. Every disk block can be used. Since there is loss of disk space due to external fragmentation, there is no need to implement disk compaction facility.
2. There is no need to determine and declare the total anticipated size of a file at the time of its creation. A file can freely grow at any time since new disk blocks required to be allocated to the file may be anywhere on the disk. 3. The method can be effectively used for both random-access and sequentialaccess files. Once the index block of a file has been loaded into the memory (which is done when the file is opened for use), the block corresponding to any part of the file can be directly accessed by using the information in index block. The method, however, has the following drawbacks: 1. It suffers from wastage of disk space especially when most files are small. This is because each file is allocated one index block irrespective of the number of data blocks allocated to the file. Hence if a file has one or two data blocks, an entire index block is used for storing just one or two pointers, leaving the index block area largely unused. One way to take care of this problem is to have enough space in the directory so that the first few (say n) block addresses can be stored in the directory itself. So for small files, all the necessary information is right in the directory. An index block is allocated for a file only when the file requires more than n blocks for storing its data. 2. It suffers from the problem of handling very large files. This is because if the index block has space that is sufficient for storing only m disk block addresses, then how can we store files whose size is larger than m disk blocks? This problem is often taken care of by using the method of multilevel index. In this method, one of the addresses in the index block is the address of a disk block called the secondlevel index block. This block contains pointers to additional disk blocks allocated to the file. This approach could be continued to a third or fourth level, but two levels of indexes are generally sufficient. DEVICE MANAGEMENT For processing data, a computer must first input data and programs for which it needs input devices. Similarly, to produce the results of processing, it needs output devices. Chapter 9 provided a description of several of these I/O devices. The device management module of an operating system takes care of controlling all the I/O devices of a computer system and provides a simple and easy to use interface to these devices. How these requirements are taken care of by an operating system is described below. Controlling I/O Devices We saw in Chapter 9 that device controllers are used for connecting I/O devices to a computer system. Each device controller is in charge of a specific type of device such as disk drives, tape drives, printers. A device controller maintains some local
buffer storage and is responsible for moving data between the I/O device it controls and its local buffer storage. Each device controller also has a few registers that it uses for communicating with the CPU. On some computers, these registers are part of the regular memory address space. This scheme is called memorymapped I/O. On other computers, a special address space is used for I/O, with each device controller allocated a certain portion of it. To perform an I/O operation, the operating system writes the relevant commands and their associated parameters into the appropriate controller's registers. After having finished loading the commands into the controller's registers, the CPU resumes with its normal operation. The device controller then examines the contents of these registers and performs necessary actions for the requested I/O operation. For example, the action for a read request will be to transfer data from the specified input device to its local buffer. Once the transfer of data from input device to the controller's local buffer is complete, one of the following two methods is used to transfer the data from the controller's local buffer to the appropriate memory area of the computer: 1. Non-DMA Transfer. In this method, as soon as the transfer of data from input device to the controller's local buffer is complete, the controller causes an interrupt to allow the operating system to gain control of the CPU. When the CPU is interrupted, it stops what it is currently doing and routine for handling the interrupt is located. On execution of the instructions of the interrupt service routine by the CPU, the data gets transferred from the local buffer of the device controller to main memory. Once this transfer of data is completed, the CPU can then resume with the computation that it was doing when it was interrupted. DMA Transfer. In the method of non-DMA transfer, the CPU is involved in the transfer of data from the device controller's buffer to the main memory. That is, the operating system executes the interrupt service routine, which reads one byte or word at a time from the device controller's buffer and stores it in memory. Involvement of the CPU in reading and storing one byte or word at a time wastes CPU time. To free the CPU from doing this low-level work, many device controllers support direct memory access (DMA) mechanism. In this method, when the operating system writes the relevant commands-and their associated parameters into the controller's registers, the command parameters also include the memory address from/to where data is to be transferred and the number of bytes to be transferred. Now after the controller has read the data from the device into its buffer, it copies the data one byte or word at a time from its buffer into the main memory at the specified memory address without the involvement of the CPU. The device controller causes an interrupt only after the entire data has been transferred to the memory. Note that in this case when the operating system starts
after the interrupt, it does not have to copy the data from the device controller's buffer to memory, since the data is already there. So it can directly continue performing the subsequent operations. Simple and Easy User Interface to I/O Devices Ease of use is an important goal of all operating systems. To meet this objective, it is important to provide simple and easy user interface to all I/O devices. This is achieved by organizing the software for using I/O devices as a series of layers, with the lower layers designed to hide the internal details of the devices from the upper layers, and the upper layers designed to present a nice, clean, uniform interface to the users. As shown in Figure 14.24, the I/O software is typically structured into the following four layers: 1. Interrupt Handlers. As mentioned above, the device controller causes an interrupt after it has completed transfer of data either from the device to its buffer (in non-DMA transfer) or from the device to memory (in DMA transfer). This interrupt is actually caused by the interrupt handler software. The interrupt handler sends an interrupt signal to the appropriate device driver to wake it up so that it can take the further course of action. 2. Device Drivers. There is a device driver for each type of device connected to the system. For example, a system normally has a disk driver for disks, a printer driver for printers, a tape driver for tapes, and so on. The device driver contains all device-dependent code. A device driver accepts abstract I/O requests from the device-independent software layer above it, and translates it into commands for the appropriate device controller. It then writes these commands and their associated parameters into the controller's registers. 3. Device-independent Software. This layer of software performs the I/O functions that are common to all devices, and provides a uniform interface to the user-level software. It takes care of mapping symbolic device names on to the appropriate device driver. Some I/O devices, such as disks, are sharable in the sense that they can be used by many users at the same time. Other devices, such as tapes and printers, have to be dedicated to a single user until that user has finished using it. The device-independent software layer handles both shared and dedicated devices in such a manner that avoids problems. 4. User-level Software. This layer of software contains the standard I/O system calls in the form of library procedures. The users use these system calls, such as read, write, print, send, and receive for performing the desired I/O operations. SECURITY
We saw towards the beginning of this chapter that security in computer systems deals with protecting the various resources and information of a computer system against destruction and unauthorized access. A total approach to computer security involves both external and internal security. External security deals with securing the computer system against external factors such as fires, floods, earthquakes, stolen disks/tapes, leaking out of stored information by a person who has access to the information, and so on. For external security, the commonly used methods include maintaining adequate backup copies of stored information at places far away from the original information, using security guards to allow the entry of only authorized persons into the computer center, allowing the access to sensitive information to only trusted employees/users, and so on. Internal security, on the other hand, mainly deals with the following aspects: 1. User authentication. Once a user is allowed physical access to the computer facility, the user's identification must be checked by the system before the user can actually use the facility. This requirement is taken care of by user authentication mechanisms. 2. Access control. A computer system contains many resources and several types of information. Obviously, not all resources and information are meant for all users. Therefore, even when a user passes the authentication phase and is allowed to use the computer facility, a way is needed to prohibit the user from accessing those resources/information that he/she is not authorized to access. This requirement is taken care of by access control mechanisms. 3. Cryptography. A third type of internal security is often used to protect unauthorized access of information that is highly critical in nature. That is, even if a user somehow manages to gain access to some information that he/she is not authorized to access, a way is needed to ensure that the user cannot make use of that information. This requirement is taken care of by cryptography mechanisms. Below we will discuss about the commonly used mechanisms for providing these three different types of internal security in computer systems. User Authentication User authentication deals with the problem of verifying the identity of a user (person or program) before permitting access to the requested resource. That is, an authentication mechanism prohibits the use of the system (or some resource of the system) by unauthorized users by verifying the identity of a user making a request. Authentication basically involves identification and verification. Identification is the process of claiming a certain identity by a user, while verification is the process of verifying the user's claimed identity. Thus, the correctness of an authentication process relies heavily on the verification procedure employed.
The three basic approaches to user authentication are as follows: 1. Proof by Knowledge. In this approach, authentication involves verifying something that can only be known by an authorized user. Authentication of a user based on the password supplied by him/her is an example of proof by knowledge. Authentication methods based on the concept of proof by knowledge are again of two types - direct demonstration method and challenge-response method. In the direct demonstration method, a user claims his/her identity by supplying information (like typing in a password) that the verifier checks against pre-stored information. On the other hand, in the challenge-response method, a user proves his or her identity by responding correctly to the challenge questions asked by the verifier. For instance, at the time of initially registering in a system as a user, the user picks a function, for example, x + 18. When the user logs in, the system randomly selects and displays a number, say 105, in which case the user must type 123 for authentication to be successful. 2. Proof by Possession. In this approach, a user proves his/her identity by producing some item that can only be possessed by an authorized user. The system is designed to verify the produced item to confirm the claimed identity. For example, a plastic card with a magnetic strip on it that has a user identifier number written on it in invisible, electronic form may be used as the item to be produced by the user. The user inserts the card in a slot meant for this purpose in the system's terminal, which then extracts the user identifier number from the card and checks to see if the card produced belongs to an authorized user. Obviously, security can be ensured only if the item to be produced is unforgeable and safely guarded. 3. Proof by Property. In this approach, the system is designed to verify the identity of a user by measuring some physical characteristics of the user that are hard to forge. The measured property must be distinguishing, that is, unique among all possible users. For example, a special device (known as a biometric device) may be attached to each terminal of the system that verifies some physical characteristic of the user, such as the person's appearance, fingerprints, hand geometry, voice, signature. In deciding the physical characteristic to be measured, an important factor to be considered is that the scheme must be phycologically acceptable to the user community. Biometric systems offer the greatest degree of confidence that a user actually is who he/she claims to be, but they are also generally the most expensive to implement. Moreover, they often have user acceptance problems because users see biometric devices as unduly intrusive. In practice, a system may use a combination of two or more of these authentication methods. 'For example, the authentication mechanism used by automated cash-dispensing machines in banks usually employs a combination of the first two approaches. That is, a user is allowed to withdraw money only if he or
she produces a valid identification card and specifies the correct password corresponding to the identification number on the card. To provide good security with password-based authentication, it is important that passwords are kept secret and passwords are chosen in such a manner that they are difficult to guess.
Access Control Once a user or a process has been authenticated, the next step in security is to devise ways to prohibit the user or the process from accessing those resources/information that he/she/it is not authorized to access. This issue is called authorization and is dealt with by using access control mechanisms. When talking about access control in computer systems, it is customary to use the following terms: 1. Objects. An object is an entity to which access must be controlled. An object may be an abstract entity, such as a process, a file, a database, a tree data structure, or a physical entity, such as a CPU, a memory segment, a printer, and a tape drive. Each object has a unique name that differentiates it from all other objects in the system. An object is referenced by its unique name. In addition, associated with each object is a "type" that determines the set of operations that may be performed on it. For example, the set of operations possible on objects belonging to the type "data file" may be Open, Close, Create, Delete, Read, and Write, whereas for objects belonging to the type "program file," the set of possible operations may be Read, Write, and Execute. 2. Subjects. A subject is an active entity whose access to objects must be controlled. That is, entities wishing to access and perform operations on objects and to which access authorizations are granted are called subjects. Examples of subjects are processes and users. 3. Protection rules. Protection rules define the possible ways in which subjects and objects are allowed to interact. That is, protection rules govern the subjects' access to objects. Therefore, associated with each (subject, object) pair is an access right that defines the subset of the set of possible operations for the object type that the subject may perform on the object. The complete set of access rights of a system defines which subjects can perform what operations on which objects. At any particular instance of time, this set defines the protection state of the system at that time. With the above mentioned concepts, access to objects by subjects is controlled in the following manner. Suppose a subject S requests for an access r on object O,
where r belongs to the set of operations that may be performed on 0. To validate this access request, the access control modules of the security system check whether access r is permitted for subject £ on object O in the current protection state of the system. If yes, the access is permitted; otherwise the access is denied. Cryptography Cryptography is a means of protecting private information against unauthorized access in those situations where it is difficult to ensure access control. The basic idea behind this security technique is that if it is not possible to ensure access control, it is better to prevent comprehension of information. Basic Concepts and Terminologies Two primitive operations employed by cryptography are encryption and decryption. Encryption (also called enciphering) is the process of transforming an intelligible information (called plaintext or cleartext) into an unintelligible form (called ciphertext). Decryption (also called deciphering) is the process of transforming thi information back from ciphertext to plaintext. Encryption is basically a mathematical function (encryption algorithm) having the following form: C = E(P, Ke) where P is the plaintext to be encrypted, Ke is an encryption key, and C is the resulting ciphertext. Decryption o C is performed by a matching function (decryption algorithm) that has the following form: P=D(C,Kli) where Kd is the decryption key. Note that the decryption function D is the inverse of the encryption function L Therefore we have D (E (P, Ke), =P To prevent the plaintext from being easily revealed, it must be possible to transform a given plaintext into a large variety of possible ciphertexts selected by a specific parameter. The keys Ke and Kd serve as this parameter. Tha is, the function parts remain the same but the keys are changed as often as necessary. The above described general structure of a cryptosystem is illustrated in Figure 14.25. Symmetric and Asymmetric Cryptosystems
There are two broad classes of cryptosystems - symmetric and asymmetric. In a symmetric cryptosystem, either both the encryption key (Ke) and decryption key (Kd) are the same or one is easily derivable from the other Usually, a common key (K) is used for both enciphering and deciphering. For security, it is important that the ke of a symmetric cryptosystem be easily alterable and must always be kept secret. This implies that the key i known only to authorized users. Symmetric cryptosystems are also known as shared-key or private-key cryptosystems. In an asymmetric cryptosystem, on the other hand, the decryption key (Kd) is not equal to the encryption key (Ke Furthermore, it is computationally impractical to derive Kd from Ke. Because of this property, only Kd needs to b kept secret and Kc is made publicly known. Asymmetric cryptosystems are also known as public-he cryptosystems. The Data Encryption Standard (DES) cryptosystem is the best known and the most widely used symmetric cryptosystem today. On the other hand, the Rivest-ShamirAdleman (RSA) cryptosystem is the first published and practically the most satisfactory asymmetric cryptosystem today. COMMAND INTERPRETATION The command interpretation module (known as command interpreter) of an operating system serves as an interface for the user to communicate with the computer via its operating system. It provides a set of commands using which the user can give instructions to the computer for getting some job done by it. The commands supported by the command interpretation module are known as system calls. When a user gives instructions to the computer by using these system calls, the command interpreter takes care of interpreting these commands and directing the system resources to handle the requests. Thus the command interpreter hides the hardware details of the system from the user. In this manner, it greatly contributes to the 'ease of use' objective of an operating system. The two broad categories of user interfaces supported by various operating systems are command-line interface and graphical user interface. They are briefly described below. Command-line Interface This is the textual user interface in which the user gives instructions to the computer by typing commands. That is, to enter a command, the user uses the keyboard to type words and symbols. For example in Unix, the user has to type "del report.txt" to delete the file named report.txt. If the User types a command incorrectly, the command interpreter will respond with a message indicating that it
did not understand the command. When this happens, the user has to just retype the command correctly. There is no problem in typing simple commands as illustrated in the example above. However, users often need to give detail instructions to the computer about their jobs when they submit them for execution. For example, while submitting a job for execution, a user may need to supply the following information: 1. His/her identification for security and accounting purposes. 2. The software and hardware resource requirements of the job. 3. The I/O device requirements of the job. 4. The directories in which the data files to be used by the job are stored. 5. The action, which the system should take in exceptional conditions, such as when the input data is missing or incorrect, or when an I/O device malfunctions. To facilitate this, systems, which support command-line interface, also support some type of command language (CL) or job-control language (JCL). Users can write codes in the JCL to give instructions to the system. These coded statements tell the operating system such things as the name of the job, the user's name and account number, the I/O devices to use during processing, the compiler to use if language translation is needed, where to find the data files, and so on. The command interpreter is designed to interpret the codes written in JCL and invoke appropriate system actions. Graphical User Interface User interface plays an important role in ease of use. The more intuitive the user interface of a system is, the easier it is to use. An intuitive user interface allows a user to use the system effectively even if he/she has never seen it before. That is, how the user interface works is very obvious to any user. Such an user interface is called user friendly. Graphical user interface or GUI (pronounced "gooey") is a user-friendly interface that is much easier to learn and use than a command-line interface. Unlike command-line interface in which commands are textual, GUI commands are graphical (pictorial). As shown in Figure 9.3 (refer to Chapter 9), a GUI provides to the user a screen full of graphic icons (small images on the screen) or menus and allows the user to make a rapid selection from the displayed icons or menus to give instructions to the computer. A point-and-draw device is normally used to rapidly point to and select a particular graphic icon or menu item from the multiple options displayed on the screen. For example, we saw that in a system that uses command-line interface, to delete a file named report.txt we need to type a command like "del report.txt.'" However, in a system that uses GUI, the same
operation can be performed simply by using a mouse to drag the icon that represents the file until it is superimposed on an icon shaped like a trash can. Then releasing the mouse button that was used for dragging the file, causes the file to disappear into the bulging trash can. Shell The command interpreter of an operating system that serves as its user interface is often referred to as the shell because it forms the outer layer of an operating system covering the other modules of the operating system. The shell can be a command-line interface or GUI. It is also possible to have many different shells for the same operating system. For example, an operating system can have a command-line shell as well as a GUI shell allowing the users to switch from one shell to another. Beginners often find it convenient to work with GUI shell, whereas advanced users find it more convenient to work with command-line shell. Some operating systems also support multiple command-line shells. It is like supporting multiple JCLs and allowing the user to use a language that he/she is most comfortable with. For example, many Unix operating systems support three command-line shells known as C shell, Bourne shell, and Korn shell. Operating System Structure We have studied about the various components of an operating system in the sections above. We will now have a look at how these components are put together to form an operating system. Layered Structure Most modern operating systems organize their components into a number of layers (levels), each built on top of lower layers. The bottom layer (layer 0) is the hardware, and the highest layer (layer n) is the user interface. The number of inbetween layers and their contents vary from one operating system to another. How this is decided is explained below. The main advantage of the layered approach is modularity. The layers are selected such that each layer uses the functions and services provided by its immediate lower layer. This approach greatly simplifies the design and implementation of the system because each layer is implemented using only those operations provided by its immediate lower level layer. Kernel
The kernel of an operating system is its central controlling part that implements the most primitive of the system's functions. It is the only part of an operating system that a user cannot replace or modify. The precise interpretation of the system's functions that are part of the kernel varies from one operating system to another. However, typical operating system kernels contain basic functions that are required for process management, memory management, device management, and low-level security features like access control. In some systems, the kernel is larger and provides for more than these functions, whereas in others, it is smaller. Monolithic Kernel versus Microkernel The two commonly used models for kernel design in operating systems are the monolithic kernel and the microkernel. In the monolithic kernel model, most operating system services such as process management, memory management, device management, file management, and security are provided by the kernel. As a result, the kernel has a large, monolithic structure. On the other hand, in the microkernel model, the main goal is to keep the kernel as small as possible. Therefore, in this model, the kernel is a very small nucleus of software that provides only the minimal facilities necessary for implementing additional operating system services. The only services provided by the kernel in this model are low-level device management, a limited amount of low-level process management, and some memory management. All other operating system services, such as file management, additional process and memory management activities, and security are implemented as user-level server process. Each server process has its lown address space and can be programmed separately. Resident and Non-Resident Operating System Modules With all the functionalities of an operating system implemented, it becomes a large software. Obviously, all the functionalities of an operating system are not needed all the time. As the main memory capacity of a system is limited, it is customary to always keep in the system's memory only a very small part of the operating system and to keep its remaining part on an on-line storage device such as hard disk. Those modules of an operating system that are always kept in the system's main memory are called resident modules and those that are kept onjlard disk are called non-resident modules. The non-resident modules are loaded into the memory on demand, that is, as and when they are needed for execution. The system kernel should not be confused with the resident modules of the operating system. The two are not necessarily the same. In fact, for most operating systems they are different. The following two criteria normally determine whether a particular operating system module should be resident: 1. Its frequency of use, and
2. Whether the system can operate at all without it. For example, file directories can be maintained on disk and loaded into the memory when required. Status information for inactive processes can similarly be swapped out on disk. In fact, the resident part of an operating system is a subset of its kernel. OTHER RELATED CONCEPTS Few other concepts related to operating systems are briefly described below.
Real-time Operating Systems For many applications, successful processing means that the application produces correct results when its processing completes. However, in real life, we often come across situations in which it is not only sufficient to obtain correct results, but the results must be produced within some specified time frame (deadline) to be useful. A few examples of such applications are: 1. An aircraft must process accelerometer data within a certain period (say every 20 milliseconds) that depends on the specifications of the aircraft. Failure to do so could cause the aircraft to go away from its right course or may even cause it to crash. 2. Failure to respond in time to an error condition in a nuclear reactor thermal power plant could result in a melt-down. 3. Failure to respond in time to an error condition in the assembly line of an automated factory could result in several product units that will have to be ultimately discarded. 4. A request for booking a ticket in a computerized railway reservation system must be processed within the passenger's perception of a reasonable time (say 3 minutes). Systems that are designed for processing such applications are called real-time systems. That is, a real-time system is a system that must satisfy the requirement of producing the desired results before a certain deadline. If it delivers the correct result, but after the deadline, the system is regarded as having failed. Thus timely production of the result of processing is as important as its correctness in a realtime system. Operating systems that are designed to handle the processing requirements of realtime systems are called real-time operating systems. The goals of CPU scheduling
in traditional timesharing computer systems are optimal throughput, optimal resource utilization, and fairness. In contrast, the main goal of CPU scheduling in real-time systems is to allow as many time-critical processes as possible to be processed in time to meet their deadlines. Network and Distributed Operating Systems Over the past few decades, advancements in microelectronic technology have resulted in the availability of fast, inexpensive computer, and advancements in communication technology have resulted in the availability of cost-effective and highly efficient computer networks. The net result of the advancements in these two technologies is that the price-performance ratio has now changed to favor the use of interconnected multiple computers in place of a single, high-speed computer. Computer systems consisting of interconnected, multiple computers are called distributed computing systems. That is, a distributed computing system is a collection of computer systems interconnected by a communication network in which each computer has its own local memory and other peripheral devices, and the communication between any two computers of the system takes place by passing messages over the communication network. The operating systems commonly used for distributed computing systems can be broadly classified into two types - network operating systems and distributed operating systems. The three most important features commonly used to differentiate between these two types of operating systems are system image, autonomy, and fault tolerance capability. These features are explained below. 1. System Image. The most important feature used to differentiate between the two types of operating systems is the image of the distributed computing system from the point of view of its users. In case of a network operating system, the users view the distributed computing system as a collection of distinct machines connected by a communication subsystem. That is, the users are aware of the fact that multiple computers are being usecf. On the other hand, a distributed operating system hides the existence of multiple computers and provides a singlesystem image to its users. That is, it makes a collection of networked machines appear to its users as a virtual uniprocessor by providing similar type of user interface as provided by centralized operating systems. 2. Autonomy. In a network operating system, each computer of the distributed computing system has its own local operating system (the operating systems of different computers may be the same or different), and there is essentially no coordination at all among the computers except for the rule that when two processes of different computers communicate with each other, they must use a mutually agreed on communication protocol. Each computer functions independently of other computers in the sense that each one makes independent
decisions about the creation and termination of their own processes and management of local resources. Notice that due to the possibility of difference in local operating systems, the system calls for different computers of the same distributed computing system may be different in this case. On the other hand, with a distributed operating system, there is a single systemwide operating system and each computer of the distributed computing system runs a part of this global operating system. The distributed operating system tightly interweaves all the computers of the distributed computing system in the sense that they work in close cooperation with each other for the efficient and effective utilization of the various resources of the system. That is, processes and several resources are managed globally (some resources are managed locally). Moreover, there is a single set of globally valid system calls available on all computers of the distributed computing system. 3. Fault tolerance capability. A network operating system provides little or no fault tolerance capability in the sense that if 10% of the machines of the entire distributed computing system are down at any moment, at least 10% of the users are unable to continue with their work. On the other hand, with a distributed operating system, most of the users are normally unaffected by the failed machines and can continue to perform their work normally, with only a 10% loss in performance of the entire distributed computing system. Therefore, the fault tolerance capability of a distributed operating system is usually very high as compared to that of a network operating system. In short, both network operating systems and distributed operating systems deal with multiple computers interconnected together by a communication network. In case of a network operating system the user views the system as a collection of distinct computers, but in case of a distributed operating system the user views the system as a "virtual uniprocessor". Deadlock There are several resources in a system for which the resource allocation policy must ensure exclusive access by a process. For example, a printer needs to be exclusively allocated to a process till the process finishes using it because its simultaneous use by two or more processes will lead to garbled printed output. Since a system consists of a finite number of units of each resource type (for example, three printers, six tape drives, four disk drives, two CPUs, etc.), multiple concurrent processes normally have to compete to use a resource. In this situation, the sequence of events required to use a resource by a process is as follows: 1. Request. The process first makes a request for the resource. If the requested resource is not available, possibly because it is being used by another
process, the requesting process must wait until the requested resource is allocated to it by the system. 2. Allocate. The system allocates the resource to the requesting process as soon as possible. It maintains a table in which it records whether each resource is free or allocated and, if it is allocated, to which process. If the requested resource is currently allocated to another process, the requesting process is "added to a queue of processes waiting for this resource. Once the system allocates the resource to the requesting process, that process can exclusively use the resource by operating on it. 3. Release. After the process has finished using the allocated resource, it releases the resource to the system. The system table records are updated at the time of allocation and release to reflect the current status of availability of resources. The request and release of resources are system calls, such as request and release for devices, open and close for files, and allocate and free for memory space. Notice that of the three operations, allocate is the only operation that the system can control. The other two operations are initiated by a process. With the above mentioned pattern of request, allocation, and release of resources, if the total request made by multiple concurrent processes for resources of a certain type exceeds the amount available, some strategy is needed to order the assignment of resources in time. Care must be taken that the strategy applied cannot cause a deadlock, that is, a situation in which competing processes prevent their mutual progress even though no single one requests more resources than are available. It may happen that some of the processes that entered the waiting state (because the requested resources were not available at the time of request) will never again change state, because the resources they have requested are held by other waiting processes. This situation is called deadlock, and the processes involved are said to be deadlocked. Hence, deadlock is the state of permanent blocking of a set of processes each of which is waiting for an event that only another process in the set can cause. All the processes in the set block permanently because all the processes are waiting and hence none of them will ever cause any of the events that could wake up any of the other members of the set. A deadlock situation can be best explained with the help of an example (see Figure 14.26). Suppose that a system has two tape drives T } and T2 and the resource allocation strategy is such that a requested resource is immediately allocated to the requesting process if the resource is free. Also suppose that two concurrent processes P1 and P2 make requests for the tape drives in the following order: 1. P1 requests for one tape drive and the system allocates Tj to it. 2. P2 requests for one tape drive and the system allocates T2 to it. 3. P1 requests for one more tape drive and enters a waiting state because no tape drive is presently available.
4. P2 requests for one more tape drive and it also enters a waiting state because no tape drive is presently available. From now on, Pi and P2 will wait for each other indefinitely, since Pi will not release 7) until it gets T2 to carry out its designated task, that is, not until P2 has released T2, whereas P2 will not release T2 until it gets 7). Therefore, the two processes are in a state of deadlock. Note that the requests made by the two processes are totally legal because each is requesting for only two tape drives, which is the total number of tape drives available in the system. However, the deadlock problem occurs because the total requests of both processes exceed the total number of units for the tape drive and the resource allocation policy is such that it immediately allocates a resource on request if the resource is free. Spooling Spooling stands for Simultaneous Peripheral Operation On-Line. It is a mechanism to deal with the following issues: 1. To reduce the speed mismatch between slow I/O devices and CPU for better utilization of the-CPU. 2. To convert mutually exclusive I/O devices into non-mutually exclusive I/O devices so that they can be simultaneously used by multiple processes in a multiprogramming system. How spooling helps in dealing with these issues is explained below. Dealing with Speed Mismatch The speed mismatch between a slow I/O device such as a printer and the CPU is normally of the order of few thousand. Thus, while a slow I/O device is inputting/outputting a few bytes to/from the main memory, the CPU can perform several thousands of operations. As a result, during reading/writing of data from/to memory, the CPU is idle for a large percentage of I/O time. Spooling reduces the idle time of the CPU by placing all data that comes from an input device or goes to an output device on a magnetic disk (see Figure 14.27). The primary reason for doing this is to keep the program and data readily available to the fast and expensive CPU on a high speed I/O medium such as a disk. Reading/writing of data from/to a disk is much faster than reading/writing from/to a slow I/O device resulting in considerable reduction in CPU idle time during I/O operations involving slow devices. In most computer systems, special low-cost I/O processors are used for spooling the input data from a slow input device on to the disk or for outputting the spooled output data from the disk on to the slow output device. These I/O processors function independent of the main processor (CPU). This enables the main highspeed, expensive CPU to be fully devoted to main computing jobs. The process of
spooling is transparent to the user programs. In general, spooling makes better use of both the main memory and the CPU. Dealing with Mutually Exclusive I/O Devices There are several I/O devices that have to be used in dedicated mode for correct system operation. Such devices are called mutually exclusive devices. Printer is a typical example of such a device. For effective utilization of such devices, an operating system often converts them into non-mutually exclusive virtual devices by using spooling. For this, the operating system creates a special process, called a daemon, and a special directory, called a spooling directory. Now when a process makes a request to use the printer, instead of allocating the printer to the process, the operating system opens a file for the process's output data in the spooling directory, which is stored on the disk. All output data to be printed on the printer by the process is written in this file. The file ii closed when the process completes printing. The daemon then actually prints the file from the disk on to the printer. The daemon is the only process having permission to print the files in the spooling directory on to the printer. By protecting the printer against direct use by the users, the problem of having someone keeping it open unnecessarily long is eliminated. Printer is not the only device that can benefit from spooling in this manner. For example, file transfer over a network often uses a network daemon. To send a file somewhere, the system puts it in a network spooling directory on the disk. Later on, the network daemon takes it out from the directory and transmits it. Note that in both the examples discussed above, spooling allows a dedicated device such as printer and network to be simultaneously shared by multiple users. This is because spooling being transparent to users, users get an impression that they are directly inputting/outputting their data from/to the dedicated device, not via the daemon process. OS CAPABILITY ENHANCEMENT SOFTWARE There are several tasks of a routine nature that many users often need to perform on their computer system. For example, program compilation, sorting of file contents in a particular order, and taking backup of important files are tasks of such nature. Not all computer users can write their own programs for performing these tasks. Even if they could write, it would clearly be wasteful if each user spent a lot of time developing programs for these tasks. This gap between the functionality of an operating system and the needs of users for these frequently used routines created a market for a special category of software that enhance the capability of an operating system. These software are either supplied by the operating system vendor, or third-party software vendors, or may be a
combination of both these sources. They are normally grouped into three categories - translating programs, library programs, and utility programs. They are briefly described below. Translating Programs Translating programs, also known as language processors, are system programs that translate a source program written by the user to an object program, which is meaningful to the hardware of the computer. These include the assembler and the various compilers and interpreters available with the system. Often the program development tools used for testing and debugging of programs are also included in this category of software. Library Programs Library programs consist of frequently used functions and operations. For example, in the area of scientific applications, the commonly used library programs include those that compute mathematical functions such as sine, cosine, square root, exponential, and logarithm of numbers; and those that perform mathematical operations such as matrix multiplication, matrix inversion, statistical analysis, and conversion of numbers from one base to another (such as binary to decimal and decimal to binary). Similarly, in the area of string processing, the commonly used library programs include those that perform such operations as string comparison, string copy, string concatenation, and calculation of the length of a given string. Library programs commonly used for file manipulation include those that perform such operations as sorting the file contents to arrange its data in a specified sequence, merging the contents of two or more sorted files into one file containing all the items of all the original files in sorted order, concatenating of two files to create a single file out of their contents, searching for some specific data from a file, and file editing. Like this, there can be a very large number of library programs available in a system. A librarian program usually controls the storage and use of the library programs in the system library. It maintains a program directory for this purpose and also facilitates the addition of new programs- to the library, or deletion of unwanted or old programs from the library whose new, updated versions might have been added to the library. Utility Programs Utility programs assist the users with system maintenance tasks such as disk formatting, disk compaction, removing file fragmentation problem, data compression, data backup, and scanning the system for computer viruses. Few of the frequently used utility programs are briefly described below.
Disk Formatting Utility In the section on file management we saw that different operating systems organize and use the available storage space on a disk in different ways. For example, some use contiguous allocation, some use linked allocation, and others use indexed allocation. Moreover, the block size may be different for different operating systems, and certain areas of the disk may be reserved by the operating system for storing system-related information (not for user files). Hence, whenever a new disk (which may be a hard disk, floppy disk, or optical disk) is to be used with a system, it must be formatted properly according to the requirements of the associated operating system. The disk formatting utility allows a user to perform this task with great ease. Disk Compaction Utility This utility is useful for those systems that use the method of contiguous allocation of disk blocks to files. We saw that in this method, almost 30% of the disk space may remain unused due to external fragmentation problem after the system has been in continuous use for some time. Disk compaction utility allows a user to perform disk compaction, which moves files to bring all free blocks together. Disk compaction is a time consuming operation and may take several hours. Hence it is usually performed at night when the system is otherwise idle. Data Compression Utility This utility uses a compression algorithm to transform files into a fraction of their normal size so that they occupy less storage space on disk or can be transferred across a network in less time. The utility has an associated decompression algorithm that is used to bring back a compressed file into its normal form and size when accessed by a user. Several modem manufacturers build data-compression techniques into their modems to reduce data-transfer times. Similarly, several disk drives have data-compression techniques built into them to allow more files to be stored in the available disk space. The compression and decompression of data happens automatically that is transparent to the users. The time required to compress and decompress the data is so small that it is not noticeable. Often the amount of space saved by using this technique is significant - it can effectively double the capacity of a disk. Data Backup Utility Data stored in a computer system can be damaged or lost in several ways such as a disk crash, a virus, a hardware malfunction, or simply an accidental erasure by its
owner. So it is always suggested to keep backup of important data. A backup utility is used to create copy of data on some storage media such as floppy disk, CD-ROM, or magnetic tape that can be stored off-line away from the computer system. When some data stored on-line gets damaged or lost, it can be restored from the backup media. It is not necessary that the backup media is always kept off-line. In today's era of computer networks, some backup devices are kept online at a place away from the main computer system and the data backup is done automatically by the computer system on this device.
Antivirus Utility A computer virus is a piece of code attached to a legitimate program that, when executed, infects other programs in the system by replicating and attaching itself to them. In addition to this replicating effect, a virus normally does some other damage to the system, such as corrupting/erasing files. Therefore, due to its spreading nature, a virus can cause severe damage to a system. A typical virus works as follows. The intruder writes a new program that performs some interesting or useful function (such as some game or utility) and attaches the virus to it in such a way that when the program is executed the viral code also gets executed. The intruder now sends this infected program by mail to other users of the system or offers it for free or for a nominal charge on floppy disks. Now if anyone uses the infected program, its viral code gets executed. When the viral code of the infected program executes, it randomly selects an executable file on the hard disk and checks to see if it is already infected. Most viruses include a string of characters that acts as a marker showing that the program has been infected. If the selected file is already infected, the virus selects another executable file. When an uninfected program is found, the virus infects it by attaching a copy of itself to the end of that program and replacing the first instruction of the program with a jump to the viral code. When the viral code is finished executing, it executes the instruction that had previously been first and then jumps to the second instruction so that the program now performs its intended function. Notice that a virus spreads because every time an infected program is executed, it tries to infect more programs. Also notice that a virus does not infect an already infected file in order to prevent an object file from growing ever longer This allows the virus to infect many programs without noticeably increasing disk space usage. When a computer system suffers from virus infection, it has to be cured. Antivirus utilities are often used to cure a computer from virus infection. These utilities first identify the virus type with which the computer is infected by matching its marker against the markers of well-known viruses. Once the type is known, the original
programs are restored from their infected versions by applying a detailed knowledge of the infection method used by the virus. For example, in viruses that modify jump instructions at the beginning of the host program, recovering can be done simply by restoring the original jump to the start of the host program code. However, notice that these utilities can only cure specific known viruses. They cannot cure a newly encountered type of virus. A good antivirus utility can normally cure several hundred types of viruses and its power can be regularly improved by frequently updating it as new viruses are discovered.
SOME POPULAR OPERATING SYSTEMS In this section we will have a look at some real operating systems that are in popular use. The operating systems described are Unix, DOS, Windows, Windows NT, and Linux. Many of these systems are continuously evolving. Therefore new features keep getting added in their new versions. Hence instead of presenting a detailed description of the features of each of these systems, only their overall characteristics and some of their most interesting aspects have been presented. Unix Unix is a multi-user, time-sharing operating system. It is a very popular operating system and can be used on a wide variety of computers ranging from notebook computers to super computers. Although it can be used on a wide variety of systems, it is especially prevalent on RISC workstations such as those from Sun Mircosystems, Hewlett-Packard, IBM, and Silicon Graphics. Development of Unix Unix was developed in the early 1970s at Bell Laboratories by Ken Thompson and Dennis Ritchie for a small PDP-11 computer. It was the first operating system to be written in a high-level language C. The normal practice till then was to use assembly language for writing operating systems due to which operating systems were system dependent. That is, they were usable only on the system for which they were developed. However, since Unix was written in C language, moving it to a new machine, known as porting it, was much easier. This was an important reason for its large popularity and availability on a wide variety of systems. BSD Unix and Other Unix Versions The ease with which Unix could be modified led to the initiation of Unix-based development activities at many universities and research laboratories. The most
influential of the non-AT&T development groups was the group at the University of California at Berkeley. This group ported Unix to the VAX computer and produced the Berkeley Software Distributions (BSD) versions of Unix. Among the improvements made by this group in t original Unix version, the notable ones were support for virtual memory, demand paging, a new shell (called csh), and the TCP/IP network protocols. BSD version also added several OS capability enhancement software to Unix, including a new text editor (called vi editor), Pascal and Lisp compilers, and many more. Over the years, as Unix grew in popularity, many computer vendors have ported it on their systems and have enhanced and maintained their own versions of Unix. This has resulted in the creation of a wide variety of Unix, and Unix-like operating systems. For example, Unix supported on IBM systems is called AIX, Unix supported systems supplied by Sun Microsystems is called Solaris, Unix supported on systems supplied by DEC (Digital Equipment Coporation) is called Ultrix, and Unix supported on systems supplied by HP (Hewelett Packard) is called HP-Unix. Structure of Unix The Unix operating system is structured in the following three layers: 1. Kernel. This layer has all the modules for process, memory, file, device, and security management. 2. Shell. This layer has the command interpreter. Unix basically provides a command-line interface (some vendors have added GUI to their new versions of Unix). There are a large number of very useful commands provided by the shell. Every command is designed to do a single, very specific task. To do larger, more complex tasks, users can easily combine several commands in the form of a shell script to produce exactly the result they want. The shell script can even be stored in a file, which can be made an executable file, which when executed will act like a new shell command. An interesting service provided by the shell layer is called pipe, which allows users to send the output of one program to the input of another program without storing the output of the first program in a temporary file. For example, entering: who/sort invokes the command who, which displays the names of all currently logged in users, and pipes its output to sort. The result is a sorted list of currently logged in users. A pipeline is a connection of two or more programs through pipes. The length of a pipeline can be arbitrarily long, and the programs used in forming pipelines are called filter programs (or just filters) because they act as data filters, transforming the data passed on to the next program in the pipeline. 3. Utilities. This layer has all the OS capability enhancement software including language compilers, text editors, text processing programs, and a wide variety of utilities and tools. Over the years, a rich set of very powerful tools has been added
to this layer, which allows effective program development and system management. Because of this, Unix is often referred to by many of its users as a complete programming environment, not just an operating system. MS-DOS MS-DOS stands for Microsoft Disk Operating System. It is a single-user operating system for IBM and IBM-compatible personal computers. It was introduced in 1981 jointly by Microsoft and IBM and was the most popular operating system for personal computers in the 1980s. Its popularity started reducing in the 1990s with the launch of Microsoft Windows operating system. Structure of MS-DOS MS-DOS is structured in the following three layers: 1. BIOS. BIOS stands for Basic Input Output System. It contains device drivers for the standard devices such as keyboard, disk, floppy, printer, and display monitor. It also contains the basic low-level services such as time-of-day and system configuration analysis. Having the hardware-dependent services in BIOS ensures that the other higher-level operating system services can be offered in a hardware-independent manner, enabling those operating system services to be easily portable to other machines. The BIOS is usually supplied by the computer manufacturer, rather than by Microsoft, and is generally located, in part, in a ROM, which is a non-erasable memory. When the computer's power is switched on, control is transferred by hardware to the bootstrap procedure of the BIOS in ROM. The bootstrap procedure carries out some hardware tests to check whether the memory and other hardware devices are functioning properly. If the hardware test passes successfully, the bootstrap procedure loads into the memory a small portion of the operating system from the disk. This part of the operating system then loads the rest of the memory resident portions of the operating system into memory from the disk. This step-by-step process of loading the operating system into memory is known as booting the system. 2. Kernel. This layer contains all the modules for process management, memory management, and file management. These modules are written in machineindependent manner so that they can be easily ported to other machines. 3. Shell. This layer has the command interpreter. MS-DOS provides a command-line interface. The MS-DOS shell, known as command.com, has about 70 commands that are partitioned into two categories - internal commands and external commands. Internal commands are those that are always memory resident, whereas external commands are those, which normally reside on disk and are loaded into the memory whenever they have to be executed. Obviously, external commands are executed much more slowly than internal ones. Later a
GUI known as DOS shell was also added to this layer, which was largely a menudriven user interface. To give an idea of what the shell commands look like and what they do, some of the shell commands are given in Figure 14.28. Both the MS-DOS and Unix versions have been given to show that the commands names normally signify their task so that they can be easily remembered and used. Moreover, note that the commands are usually short so that you have to type only a few characters to invoke a task. Most of these commands require arguments, which are specific pieces of information required to complete the command. For example, the del command requires the name of the file you want to delete. Limitations of MS-DOS In spite of its wide popularity in the 1980s, MS-DOS could not continue to be popular like Unix for a long period. The following are some of the reasons often quoted for its failure: 1. It was a single user, single task operating system (did not support multitasking). That is, it assumed that only one user uses the machine and runs only one program at a time. 2. Since it was a single user, single task operating system, it did not have any protection features to protect two or more programs stored in memory from one another. 3. It had lintited program addressability to 640 Kbytes. That is, it was designed, to support programs whose maximum size was 640 Kbytes. At first, this was not a problem, since those days machines had only 64 Kbytes of RAM in it, but when later models came out with up to 16 megabytes, the inability to run programs larger than 640 Kbytes became a major limitation. 4. It was designed for the 16-bit CPUs that Intel was making in the early and middle 1980s. It could not take advantage of the 32-bit CPUs that were later made by Intel. 5. It restricted the user to eight-character file names with three-character extensions. Due to the lack of flexibility of using long file names, users often found it inconvenient to choose appropriate file names. No major operating system places this restriction on its users. 6. It had a command-line interface that some people found difficult to use and hard to learn. i Command Name Task Description In MS-DOS In Unix copy cp Copy a file del or erase rm Delete/Erase/Remove a file rename or ren mv Rename a file or move it from its old name to new name
dir Is Display a directory listing mkdir or md mkdir Create/Make a new directory chdir or cd cd Change the current working directory rmdir rmdir Remove an empty directory date date Display or change the current date type cat Display (type/catenate) contents of a file Figure 14.28. Illustrating some of the shell commands.
Microsoft Windows Microsoft Windows operating system was developed by Microsoft to overcome the limitations of its own MS-DOS operating system. The first successful version of this operating system was Windows 3.0, which was released in 1990. The subsequently released versions were Windows 95, Windows 98, and Windows 2000. The numbers associated with these released versions indicate their year of release. The main features of Microsoft Windows are as follows: 1. Its native interface is a GUI. Hence it is easier for a new user to learn and use the system. 2. Microsoft Windows was designed to be not just an operating system but also a complete operating environment. That is, Microsoft Windows programs conform to a standard way of working. For example, a Microsoft Windows word processor works similarly the way a Microsoft Windows spreadsheet (or any other type of Windows program) works. This means that the experience gained by learning one Windows program is useful while using any other Microsoft Windows program. 3. It is a single-user, multitasking operating system. That is, a user may run more than one program at a time. For example, while editing a file in the foreground, a sorting job can be given in the background. The monitor's screen can be partitioned into multiple windows and the progress of different programs can be viewed on different windows. 4. It allows file names to contain up to 255 characters as well as some punctuation marks, such as periods, commas, and semicolons. With these features, Microsoft Windows became an operating system of choice for most personal computer users after 1990. However, good GUIs require powerful processors, large primary memory, large hard disks, and a fast, high-resolution monitor. Hence Microsoft Windows requires all these on your computer for successful installation and for running most Windows applications. The actual system configuration required varies from one Windows version to another and also depends on the application. Microsoft Windows NT
Microsoft Windows NT is a multi-user, timesharing operating system developed by Microsoft. It was designed to have Unix-like features so that it can be used for powerful workstations and network and database servers. Its main features are as follows: 1. Unlike Unix, its native interface is a GUI. The look and feel of Microsoft Windows NT's GUI is similar to that of Microsoft Windows's GUI. 2. It supports multiprogramming and is also designed to take advantage of multiprocessing on systems having multiple processors. 3. It has built-in networking and communications features so that any computer with Microsoft Windows NT can be made to work as a network client or server. 4. It provides strict system security. 5. It has a rich set of tools for software development and system administration. 6. It can run Microsoft Windows applications and many Unix applications directly. 7. Its design is based on a microkernel so that it can be easily ported to many different types of machines, and its features can be easily enhanced by users. 8. It is a true 32-bit operating system in the sense that it can make full use of the 32-bit architectures of the processors and the memory and I/O bus to provide fast processing capability. Linux Linux is an open-source operating system enhanced and backed by thousands of programmers world-wide. It is a multi-tasking operating system that was originally designed to be used on personal computers. The name "Linux" is derived from its inventor Linus Torvalds. Linus was a student at the University of Helsinki, Finland in early 1990s when he wrote the first version of a Unix-like kernel as a toy project. He later posted the code on the Internet and asked programmers across the world to help him build it into a working system. The result was Linux. Torvalds holds the copyright, but permits free distribution of the source code. That is, he oversees development of the kernel and owns its trademark. When someone submits a change or a feature, Torvalds and his core team of kernel developers review the merit of adding it to the source code. Linux Development Process Linux being an open-source operating system, various parties are involved in its development process. The key roles played by the various participants are as follows: 1. The open-source development community (programmers across the world) develops the source code for the operating system kernel and submits it to the core team of kernel developers headed by Torvalds. This team reviews the merit of
adding the submitted change or feature in the kernel and incorporates it into the kernel if so decided. This is how the kernel evolves and matures to become better day-by-day. 2. Application vendors and programmers download the freely available kernel and develop all typds of application software for it. 3. Distribution companies package and sell various Linux versions. 4. Hardware vendors bundle Linux on their systems and supply their systems along with the Linux operating system and a set of application software. 5. End users run Linux and applications to meet their processing needs. Linux Distribution Versions The way Linux is developed and distributed, has caused the operating system to fragment into multiple commercial flavors. Today there are close to 200 Linux distribution versions. Some of the most popular distribution versions include Caldera OpenLinux eServer, Red Hat, SuSE, TurboLinux Server, LinuxMandrake, Connectiva, and Red Flag. These versions are available from commercial vendors who make money by selling their implementation, as well as installation, maintenance, and other services. Linux is also distributed by the nonprofit Debian organization, whfch is developing a free operating system and related software. The existence of multiple Linux distribution versions frees users from being locked into one vendor. It also creates healthy competition, which furthers innovation. Additionally, the different versions of Linux have always catered to slightly different markets. For example, the SuSE version is for users in banking and telecommunications industries, and the Connectiva version has been focused towards South American market. The main differences among various distribution versions include the set of tools, library version, number and quality of applications, documentation, support, and service. Linux Distribution and Installation Linux has traditionally been organized as a series of floppy disks - the A series containing the basic system, the D series containing the compilers and program development tools, the N series containing the modules for networking features, and so on. The total series comes in about 50 floppy disks. However, not every user needs the entire set of floppy disks. For example, you do not need the N series floppy disks, if you do not want networking features. Linux can be installed on a computer system either as the only operating system or along with other popular operating systems such as MS-DOS or Microsoft Windows. When multiple operating systems are installed on the same machine,
you can boot the machine by choosing the appropriate operating system at the time of booting. The Linux boot manager, Linux Loader (LiLo) is used for this purpose. If you are working with MS-DOS or Microsoft Windows, you do not even need to reboot in order to start Linux, you can simply use the command LOADLIN to start Linux directly from the MS-DOS command-line interface. Advantages and Limitations of Linux The following advantages are often quoted in favour of Linux: 1. One of the biggest advantages of Linux is its price. The system, as such, is free, and can be copied and distributed without restrictions. 2. It has a rich set of features, which continues to grow rapidly because thousands of enthusiastic developers continuously keep extending and improving its features. 3. For those familiar with Unix, which is one of the oldest and the most popular operating system, learning and using Linux is a simple matter because many of the Unix features and tools are supported on Linux. |Critics of Linux, however, quote that it has the following limitations: 1. The biggest obstacle to implementing Linux is the perceived lack of commercial support and service. Most enterprises are reluctant to consider Linux today due to lack of vendor commitment and a full framework of global service and support. 2. As of today, Linux offers limited support for databases containing more than a terabyte of data. 3. As of today, there are relatively few commercial Linux-based applications because of commercial vendors' misconceptions about the open source community. They think that they have to give the product away or publish their source code. However, more and more commercial vendors are gradually realizing that developing software for Linux is like developing for other platforms. 4. Another reason for the availability of relatively few commercial Linux-based applications is the existence of so many Linux distribution versions. As of today, there is a lack of standard among various Linux distribution versions for such things as what are the file formats to be used, how and where to keep the system files, how and where to add new programs to menus, etc. Due to this, a commercial software developed for Linux must be tested on every version of Linux it runs and on every version of the libraries and services it requires. To enhance application portability to the various Linux distribution versions, most commercial Linux software vendors and distributors are participating in developing a base set of libraries, application programming interfaces, a filesystem hierarchy standard, and interoperability measures called the Linux Standard Base (LSB). Points to Remember
1. An operating system (often referred to as OS) is an integrated set of programs that controls the resources (the CPU, memory, I/O devices, etc.) of a computer system and provides its users with an interface or virtual machine that is more convenient to use than the bare machine. 2. The main functions provided by most modern operating systems are process management, memory management, file management, device management, security, and command interpretation. 3. The efficiency of an operating system and the overall performance of a computer system are usually measured in terms of its throughput, turnaround time, and response time. 4. Throughput is the amount of work that the system is able to do per unit time. It is measured as the number of processes that are completed by the system per unit time. 5. Turnaround time is the interval from the time of submission of a job to the system for processing to the time of completion of the job. 6. Response time is the interval from the time of submission of a job to the system for processing to the time the first response for the job is produced by the system. 7. A process (also called job) is a program in execution. The main objective of the process management module of an operating system is to manage the processes submitted to the system in such a manner so as to minimize the idle time of the various processors (CPU, I/O processors, etc.) of the computer system. 8. In the batch processing method of job execution, job-to-job transition is automatic. This helps in reducing the idle time of a computer system to a great extent because transition from one job to another does not require any operator intervention. 9. Control statements are used by the operating system to identify a new job and to determine what resources are needed by the job during its execution. These control statements are written in a language known as the job control language (JCL). 10. In a uniprogramming system only one job is processed by the system at a time and all the system resources are exclusively available for the job until it completes. 11. CPU-bound jobs mostly perform numerical calculations, with little I/O operations. They heavily utilize the CPU during the course of their processing. 12. I/O-bound jobs normally input vast amount of data, perform very little computation, and output large amount of information. Their CPU utilization is very low. 13. Multiprogramming is the name given to the interleaved execution of two or more different and independent programs by the same computer. 14. In some multiprogramming systems, only a fixed number of jobs can be processed concurrently (multiprogramming with fixed tasks) (MFT), while in
others the number of jobs can vary (multiprogramming with variable tasks) (MVT). 15. In a multiprogramming system, the area occupied by each job residing simultaneously in the main memory is known as a memory partition. 16. The part of the operating systerrKhat is concerned with deciding which of the ready jobs should be allotted the CPU for execution is called the CPU scheduler, and the algorithm it uses is called the CPU scheduling algorithm. 17. Multitasking is the concurrent execution of multiple jobs (often referred to as tasks of same user) in a single-user system. 18. In operating systems with threads facility, a process consists of an address space and one or more threads of control. Each thread of a process has its own program counter, its own register states, and its own stack. But all the threads of a process share the same address space. Threads are often referred to as lightweight processes and traditional processes are referred to as heavyweight processes. 19. The term multiprocessing is used to describe interconnected computer configurations or computers with two or more CPUs that have the ability to simultaneously execute several programs. In such a system, instructions from different and independent programs can be processed simultaneously by different CPUs or the CPUs may simultaneously execute different instructions from the same program. 20. Time-sharing is a mechanism to provide simultaneous interactive use of a computer system by many users in such a way that each user is given the impression that he/she has his/her own computer. It uses multiprogramming with a special CPU scheduling algorithm to achieve this. 21. The memory management module of an operating system manages the main memory of the system. Its job is to keep track of which parts of memory are in use and which parts are not in use, to allocate memory to processes when they need it and deallocate when they are done. 22. The uniprogramming memory model is used in uniprogramming systems in which only one job is processed by the system at a time and all the system resources are exclusively available for the job till it completes. 23. In a multiprogramming memory model, multiple user processes can simultaneously reside in the main memory. The two memory management schemes used to facilitate this are multiprogramming with fixed number of memory partitions and multiprogramming with variable number of memory partitions. In the former scheme, the user area of the memory is divided into a number of fixed-sized partitions. On the other hand, in the latter scheme, the number, size, and location of the partitions vary dynamically as processes come and go. 24. External fragmentation of memory is a situation when enough total free memory space exists to satisfy the memory need of a process, but still the process cannot be loaded because the available free memory is not contiguous (is fragmented into multiple free blocks of memory).
25. Internal fragmentation is a situation when a process is allocated more memory than its actual memory requirement and the additional memory allocated to the process remains unutilized because it is neither used by the process to which it is allocated nor it can be allocated to any other process for use. 26. Paging is a memory management scheme used to solve the problem of external fragmentation of memory. It does this by allowing a process's memory to be noncontiguous, thus allowing a process to be allocated physical memory wherever free memory blocks are available. 27. Virtual memory is a memory management scheme that allows the execution of processes that might not be completely loaded in the main memory. The three basic concepts used for the realization of virtual memory are on-line secondary storage, swapping and demand paging. 28. A file is a collection of related information. Every file has a name, its data, and attributes. 29. The file management module of an operating system takes care of file-related activities such as structuring, accessing, naming, sharing, and protection of files. 30. Directories are used by operating systems to map file names to their corresponding file attributes and file data, and also to provide greater flexibility to users in file naming. 31. Disk space allocation methods deal with how to allocate space to multiple files on disk. The three commonly used methods for disk space allocation are contiguous, linked, and indexed. 32. The device management module of an operating system takes care of controlling all the I/O devices of a computer system and provides a simple and easy to use interface to these devices. 33. Security in computer systems deals with protecting the various resources and information of a computer system against destruction and unauthorized access. 34. External security deals with securing the computer system against external factors such as fires, floods, earthquakes, stolen disks/tapes, leaking out of stored information by a person who has access to the information, and so on. 35. Internal security mainly deals with the mechanisms for user authentication, access control, and cryptography. 36. User authentication deals with the problem of verifying the identity of a user (person or program) before permitting access to the requested resource. The three basic approaches to user authentication are proof by knowledge, proof by possession, and proof by property. 37. Access control mechanisms are used to prohibit a user/process from accessing those resources/information that he/she/it is not authorized to access. 38. Cryptography is a means of protecting private information against unauthorized access in those situations where it is difficult to ensure access control. The basic idea behind this security technique is that if it is not possible to ensure access control, it is better to prevent comprehension of information.
39. The command interpretation module (known as command interpreter) of an operating system serves as an interface for the user to communicate with the computer via its operating system. The two broad categories of user interfaces supported by various operating systems are command-line interface and graphical user interface. 40. The kernel of an operating system is its central controlling part that implements the most primitive of the system's functions. It is the only part of an operating system that a user cannot replace or modify. 41. Those modules of an operating system that are always kept in the system's main memory are called resident modules and those that are kept on hard disk are called non-resident modules. The non-resident modules are loaded into the memory on demand, that is, as and when they are needed for execution. 42. A real-time system is a system that must satisfy the requirement of producing the desired results before a certain deadline. Operating systems that are designed to handle the processing requirements of real-time systems are called real-time operating systems. 43. Computer systems consisting of interconnected, multiple computers are called distributed computing systems. 44. The operating systems commonly used for distributed computing systems can be broadly classified into two types - network operating systems and distributed operating systems. In case of a network operating system the user views the system as a collection of distinct computers, but in case of a distributed operating system the user views the system as a "virtual uniprocessor". 45. Deadlock is the state of permanent blocking of a set of processes each of which is waiting for an event that only another process in the set can cause. All the processes in the set block permanently because all the processes are waiting and hence none of them will ever cause any of the events that could wake up any of the other members of the set. 46. Spooling (Simultaneous Peripheral Operation On-Line) is a mechanism to reduce the speed mismatch between slow I/O devices and CPU, and to convert mutually exclusive I/O devices into non-mutually exclusive I/O devices. 47. The gap between the functionality of an operating system and the needs of users for frequently used routines is often bridged by a special category of software known as OS capability enhancement software. These software are normally grouped into three categories - translating programs, library programs, and utility programs. 48. Unix is a multi-user, time-sharing operating system. It is a very popular operating system and can be used on a wide variety of computers ranging from notebook computers to super computers. 49. MS-DOS stands for Microsoft Disk Operating System. It is a single-user operating system for IBM and IBM-compatible personal computers. It was introduced in 1981 jointly by Microsoft and IBM and was the most popular operating system for personal computers in the 1980s.
50. Microsoft Windows operating system was developed by Microsoft to overcome the limitations of its own MS-DOS operating system. The first successful version of this operating system was Windows 3.0, which was released in 1990. The subsequently released versions were Windows 95, Windows 98, and Windows 2000. It became an operating system of choice for most personal computer users after 1990. 51. Microsoft Windows NT is a multi-user, timesharing operating system developed by Microsoft. It was designed to have Unix-like features so that it can be used for powerful workstations and network and database servers. 52. Linux is an open-source operating system enhanced and backed by thousands of programmers world-wide. It is a multi-tasking operating system that was originally designed to be used on personal computers. Questions 1. What is an operating system? Why is it necessary for a computer system? 2. What are the two primary objectives of having an operating system in a computer system? Explain how an operating system helps in meeting these objectives. 3. What is a 'virtual machine'? Explain how an operating system helps in converting a computer system into a virtual machine. 4. Draw the logical architecture diagram of a computer system. Explain the role of an operating system in this architecture. 5. Explain the role of an operating system as a resource manager of a computer system. 6. "The operating system tends to isolate the hardware from the users". Discuss this statement. 7. "The operating system acts as the manager of the various resources of a computer system". Discuss this statement. 8. List out the various functions normally performed by an operating system. 9. Explain the role of an operating system with respect to the following functions: (a) Process management (d) Device management (b) Memory management (e) Security (c) File management (f) Command interpretation 10. What are the main parameters normally used for measuring the performance of a computer system? 11. Define the following terms: (a) Throughput (b) Turnaround time (c) Response time 12. Differentiate among the following terms: (a) Throughput
(b) Turnaround time (c) Response time 13. What is a process in a computer system? What is the main objective of the process management module of an operating system? 14. How was a job typically executed in early computer systems? What was the major drawback in this approach of job execution? 15. What is batch processing? How are jobs typically executed in a batch processing system? 16. Explain how batch processing mechanism helped in overcoming the main drawback of manual loading mechanism used for processing jobs in early computer systems. 17. Explain how the computer separates one job from another from a batch of jobs for automatic job-to-job transition in a batch processing system. 18. What are control statements? Why are they needed? 19. Do all computers use the same types of JCL statements? Why? 20. You want to compile and execute a FORTRAN program. In plain English, list out the necessary JCL statements you will prepare for this jpb. ., 21. What is the difference between a uniprogramming system and a multiprogramming system? What are their relative advantages and disadvantages? 22. Define multiprogramming. Explain how multiprogramming ensures effective utilization of main memory and CPU. 23. Differentiate between I/O-bound and CPU-bound jobs. 24. List out some of the hardware and software facilities required for a multiprogramming system to work satisfactorily. 25. What is a memory partition? 26. Explain how CPU is switched from one job to another in a multiprogramming system and how is it ensured that a temporarily suspended job will execute correctly when the CPU is given back to it after some time. 27. What is a process control block? Why is it needed? What does it typically contain? 28. What is multitasking? What are the similarities and differences (if any) between multiprogramming and multitasking? 29. What are threads? Differentiate between a single-threaded process and a multi-threaded process. 30. What are the main motivations for using a multithreaded process instead of multiple single-threaded processes for performing some computation activities? •1. Differentiate between multiprogramming and multithreading. >2. What is multiprocessing? Draw the basic organization diagram of a typical multiprocessing system. 33. Differentiate between tightly-coupled and loosely-coupled multiprocessing systems. 34. How is multiprocessing different from multiprogramming? 35. Discuss the advantages and limitations of multiprocessing systems.
36. Can multiprocessing and multiprogramming be part of the same system? Give reasons for your answer. 37. What is time-sharing? What is a time-slice? 38. In a timesharing system, explain how each and every user feels that he/she is the sole user of the computer system. 39. What are the three different states in which all users of a timesharing system fall? Illustrate how a particular user switches from one state to another. 40. Is multiprogramming used in a time-sharing system? Give reasons for your answer. 41. List out some of the hardware and software requirements of a time-sharing system. 42. List out some of the main advantages of time-sharing systems. 43. Why are timesharing systems considered to be most suitable for program development and testing? 44. Multiprogramming and timesharing both involve multiple user processes in the computer concurrently. What is the basic difference between the two concepts? 45. Differentiate among the following terms: (a) Multiprogramming (d) Multithreading (b) Multitasking (e) Timesharing (c) Multiprocessing 46. What is the main objective of the memory management module of an operating system? 47. Differentiate between uniprogramming and multiprogramming memory models. What are their relative advantages and disadvantages? 48. Differentiate between multiprogramming with fixed number of memory partitions and multiprogramming with variable number of memory partitions. 49. What is memory fragmentation problem? Differentiate between internal and external fragmentation of memory. 50. What is internal fragmentation of memory? Illustrate with an example. 51. What is external fragmentation of memory? Illustrate with an example. 52. What is paging? Illustrate paging mechanism with the help of an example. 53. Does paging mechanism suffer from internal fragmentation or external fragmentation of memory? Give reason for your answer. 54. What is a virtual memory? Why is it so called? How is it useful? 55. Explain the three basic concepts used for the realization of virtual memory. 56. What is swapping? How does it help in memory management? 57. What is a page replacement algorithm? What is a victim page? 58. List out the advantages and disadvantages of virtual memory. 59. If virtual memory is used carelessly, it may substantially decrease performance instead of increasing performance. Discuss. 60. What is a file? Differentiate between a file's data and its attributes. 61. What is a file name? Why is it used?
62. What are the typical jobs performed by the file management module of an operating system? 63. Describe the two popular types of file structures used by operating systems. 64. Describe the two commonly supported file»0ccess methods at operating system level. 65. Differentiate between sequential access and random access files. 66.. List out some of the typical file operations provided by an operating system and mention what each operation is used for. 67. What is a file extension in a file name? What are file extensions typically used for? List out some typical file extensions and their meaning. 68. What is a directory? What does a directory typically contain? 69. Describe the three commonly used types of directory organizations. Discuss their relative advantages and disadvantages. 70. Describe the three commonly used methods for disk space allocation. Discuss their relative advantages and disadvantages. 71. What are the typical jobs performed by the device management module of an operating system? 72. What is memory-mapped I/O? 73. Explain the non-DMA and DMA methods of transferring data from a device controller's local buffer to the appropriate memory area of the computer. Which method is better and why? 74. How does an operating system provide simple and easy user interface to all I/O devices? 75. Draw a diagram showing the various layers in which the I/O software for device management is typically structured. Explain the functions performed by the software module at each of these layers. 76. What are the typical jobs performed by the security module of an operating system? 77. Differentiate between external security and internal security requirements of a computer system. 78. What are the various security aspects that an operating system normally needs to deal with as part of the internal security of a computer system? 79. Differentiate between user authentication and access control requirements of a computer system. 80. What is cryptography? How does it help in improving the security of a computer system? 81. What is user authentication? Describe the three commonly used approaches for user authentication in modern computer systems. 82. Differentiate between direct demonstration and challenge-response methods of user authentication. 83. What is a biometric device? What is it used for? Why are biometric devices not very popular?
84. What is access control? Define the following terms with respect to access control: (a) Object (b) Subject (c) Protection rules 85. Explain how a process's request for accessing an object is validated by the system. 86. What is cryptography? What is the basic idea behind this security technique? 87. Define the following terms with respect to cryptography: (a) Encryption (c) Plaintext (b) Decryption (d) Cipher text 88. Draw a diagram illustrating the general structure of a cryptosystem. Explain the roles of various components of this diagram. 89. Differentiate between symmetric (private-key) and asymmetric (public-key) cryptosystems. 90. What are the typical jobs performed by the command interpretation module of an operating system? 91. What is a command interpreter? How does it contribute to the 'ease of use' objective of an operating system? 92. What are system calls? How are they used? 93. Briefly describe the features of the two types of user interfaces commonly supported in modern operating systems. 94. Differentiate between command-line interface and graphical user interface. List out their relative advantages and disadvantages. 95. What is a job control language (JCL)? Why is it needed? 96. What is a shell? Why is it so called? Can an operating system have many different shells? Give reason for your answer. 97. Why do most modern operating systems use the layered structure for organizing their various modules? 98. What is a kernel of an operating system? What basic functions of an operating system are typically part of the kernel? 99. Differentiate between the monolithic kernel and microkernel approaches of structuring the modules of an operating system. 100. Differentiate between resident and non-resident operating system modules. What are the main criteria for deciding whether a particular operating system module should be resident or non-resident? 101. What is a real-time system? Give some examples of applications that require real-time systems for their processing. 102. What is a distributed computing system? Give an example of an application that uses a distributed computing system.
103. What are the two common types of operating systems used for distributed computing systems? What are the important features commonly used to differentiate between these two types of operating systems? 104. What is a deadlock? Give an example to illustrate a deadlock situation. 105. What is spooling? How does it help in improving the performance of a computer system? 106. Explain how spooling can be used to convert a mutually exclusive I/O device into a shared resource. 107. What are OS capability enhancement software? Name a few types of software that fall in this category. 108. What are translating programs? Name a few types of software that fall in this category. 109. What are library programs? Give a few examples of such programs. What is a librarian program? 110. What are utility programs? Give a few examples of frequently used utility programs. 111. Write short notes on: (a) Disk formatting utility (d) Data backup utility (b) Data compaction utility (e) Antivirus utility (c) Data compression utility i 12. What is a computer virus? How does a typical virus works? When a computer system suffers from virus infection, how is it cured? 113. Name four popular operating systems. Describe the main features of any two of them. 114. What was the main reason for the large popularity and availability of Unix on a wide variety of systems? 115. Describe the structure of the Unix operating system. 116. Explain the concept of pipe in Unix. What is a pipeline? 117. Describe the structure of MS-DOS. 118. What is a BIOS? Where does it normally reside? Who usually supplies it? 119. What is meant by booting of a system? Explain how an MS-DOS system is booted. 120. Differentiate between internal and external commands of MS-DOS. 121. What are some of the main reasons for the failure of MS-DOS? 122. List out some of the key features of Microsoft Windows. 123 List out some of the key features of Microsoft Windows NT. 124. List out some of the key features of Linux. Explain the development process of Linux. 125. List out some of the key limitations of Linux. 126. Write short notes on the following: (a) Objectives of an operating system (b) Functions of an operating system (c) Parameters for measuring system performance
127. Write short notes-on the fallowing: (a) Batch processing (b) Multiprogramming (c) Multitasking (d) Multiprocessing (e) Multithreading (f) Timesharing 128. Write short notes on the following: (a) Process control block (b) Deadlock problem (c) Motivations for using threads (d) Tightly-coupled and loosely-coupled multiprocessing systems (e) CPU bound and I/O bound jobs 129. Write short notes on the following: (a) Uniprogramming memory model (b) Multiprogramming memory model (c) Memory fragmentation (d) Paging mechanism (e) Virtual memory 130. Write short notes on the following: (a) File structures (b) File access methods (c) File naming (d) File operations 131. Write short notes on the following: (a) Directory contents (b) Directory organization (c) Linked allocation method of disk space (d) Indexed allocation method of disk space (e) Contiguous allocation method of disk space 132. Write short notes on the following: (a) Memory-mapped I/O (b) DMA transfer (c) Interrupt handler (d) Device driver 133. Write short notes on the following: (a) User authentication (b) Access control (c) Cryptography 134. Write short notes on the following: (a) Challenge-response method of user authentication (b) Direct demonstration method of user authentication (c) Biometric devices
(d) Symmetric and asymmetric cryptosystems 135. Write short notes on the following: (a) GUI (b) JCL (c) Shell (d) Kernel (e) Command-line interface 136. Write short notes on the following: (a) Monolithic kernel versus Microkernel (b) Resident and non-resident operating system modules 137. Write short notes on the following: (a) Spooling (b) OS capability enhancement software (c) Real-time operating systems (d) Distributed operating systems 138. Write short notes on the following: (a) Translating programs (b) Library programs (c) Utility programs (d) Computer virus
Chapter 15 Application Software Packages This chapter provides greater detail of some of the most commonly known application software packages, which were introduced in Chapter 10. Out of the ones discussed in Chapter 10, we have chosen to provide additional details of only those application software that are readily available as pre-written software packages and are most widely used packages. In particular, the details of the following application software packages have been provided: 1. Word-processing package . 2. Spreadsheet package 3. Graphic? package 4. Personal Assistance package Each of these pre-written software packages is available from multiple vendors. However, the features of a particular type of software package from different vendors are more or less same irrespective of the vendor. Only their mode of operation may vary from one vendor's package to that of another. Hence in this chapter we will mainly concentrate on the general features of the packages rather than their mode of operation. Once you are familiar with the features, the mode of operation can be easily learnt for the specific software package that you have to use by using its operation manual. WORD-PROCESSING PACKAGE What it is? Word-processing is a term that describes the use of hardware and software to create, edit, view, format, store, retrieve and print documents (written material such as letters, reports, books, etc.). A word-processing package enables us to do all these on a computer system. Commonly Supported Features Today's word-processing packages normally support the features described below. Entering Text A word-processing package allows you to enter text with the computer's keyboard. The text to be entered is typed on the keyboard. Every character typed gets immediately displayed on the computer's screen. While entering text, you need not worry about moving to the next line as your text approaches the end of the current line. The word wrap feature of word-processing software determines when the current line is full, and the text that follows is automatically moved to the next
line. Since the software recognizes when to begin a new line and automatically moves words to the next line, the only time you have to press the Enter key is at the end of a paragraph. The word-wrap feature maintains the format of the lines of a paragraph continuously and automatically. This feature is very useful for people with fast typing speeds because they can go on entering text at their speed without the need to keep track of where to end a line. Editing Text This feature allows you to make changes in an already entered document. Changes that might have required extensive retyping a few years ago with a typewriter can now be made on the screen with a few keystrokes with a word-processing package. While editing, you can use either insert mode or typeover mode. In insert mode, new characters typed get inserted in the text at the position of the cursor. That is, characters to the right of the cursor move to the right to make room for the new characters as they are typed. In typeover mode, the new characters typed replace (type over) the existing characters at the position of the cursor. While editing you can also delete unwanted text. Few characters or words at the cursor position are normally deleted with the use of Delete key (to delete characters to the right of the cursor) or Backspace key (to delete characters to the left of the cursor). However, a block of text (say a few lines, or a few sentences, or an entire paragraph or page) is deleted by first selecting (highlighting) it with the use of a mouse or a combination of keys and then deleting it all at once by pressing the Delete key. Whenever some text is deleted, the text following the deleted text is automatically moved over to fill the space generated by the deleted text. Text editing feature also supports cut-and-paste facility that allows you to move a block of text from one place to another within a document or even between documents. Sometimes a block of text may be used at several locations in a document, or in several documents. In such situations, a slightly modified version of this facility, known as copy-and-paste facility, is used in which the block of text is copied at desired positions but not deleted from its original position. Formatting Page Style This feature allows you to define the page layout format for a document. Definition of a page layout format many include things such as: 1. Space to be left for left, right, top, and bottom page margins.
2. The default spacing between two lines (such as single space, double space, 1.5 space, etc.). 3. Automatic numbering of pages with page number style (such as 1, 2, 3, ..., or 1/10, 2/10, 3/10, ..., 1 of 10, 2 of 10, 3 of 10, ..., etc.) and page number position (top center, top left and right, bottom center, bottom left and right, etc.). 4. Automatic placement of header and footer labels on each page (such as the header labels used in this book). 5. Setting text in multiple columns, one or more columns of text on a single page. The multi-column option is frequently used in newsletter and newspaper printing. It is also possible to create and store multiple standard page format styles (called style sheets) for different categories of documents like memos, letters, reports, books, etc. Most word-processing software come with a few pre-stored style sheets. The basic idea is to save time and effort of re-creating a page format every time you have to create a new document. When you create a new document, you simply choose a style sheet from the list of pre-stored style sheets depending on the document type. All the page format specifications defined in the chosen style sheet automatically apply to the new document. You can even modify a few specifications of a style sheet that you want differently from the standard specifications in the style sheet. This can be done much faster than defining all the format specifications afresh. The beauty of a word-processing package is that you can change page format specifications as often as needed. There is no need to re-type the document if it is to be printed with a different page format. For example, a document prepared with multi-column page format can be converted to a single-column page format just by changing the page format style. The document is printed according to the most recent format specifications. Formatting Text A word-processing package not only allows you to format page styles but it also has features for formatting portions of text to improve the general appearance and readability of a document. Text formatting features normally include things such as: 1. Selection of an appropriate font. A font is a complete set of characters with the same style and size. A word-processing package comes with several standard fonts such as Times, Helvetica, Palatino,etc. Different fonts may be applied to different portions of the same document. For example, chapter heading, section heading and the running text of a book may have three different fonts. A particular font should be used only if your word-processing package as well as your printer supports that font. Figure 15.1 shows a few different font types.
2. Selection of an appropriate font size. Font size is measured in points. A point is 1/72 of an inch, and the size refers to the distance from the top of the tallest character to the bottom of the character that extends the lowest (such as y, p or q). Different font sizes may be applied to different portions of the same document. For example, chapter heading, section heading and the running text of a document usually use three different font sizes. Figure 15.2 shows a few different font sizes. This sentence is written in Times New Roman font.
This sentence is written in Helvetica font. This sentence-is written in Palatino font. This sentence is written in Courier New font.
This sentence is written in Antique Olive font. 3. Selecting an appropriate font style. Commonly used font styles are italic, bold and underline. They are normally used to highlight individual words or phrases or portions of text in a document. Figure 15.3 illustrates their use in a document. This sentence is written in italic style. This sentence is written in bold style. This sentence is written in underline style. You can even make individual words italic, bold, or underline. 4. Selecting an appropriate justification. Justification is the alignment of text on the left or the right margin, or on both margins. In case of left-justification, beginnings of lines are aligned with the left margin of the pate. This style is also known as ragged right because the ends of lines at the right edge just end where the last word ends. In case of right-justification, ends of lines are aligned with both the right margin of the page. In case of full-justification all the lines are properly aligned with both the left and right margins of the page. A world-processing software accomplishes this by making micro adjustment of space between the words, and sometimes even between the characters on a line. In case of centerjustification of a particular line, it is placed at the center of a line with equal spacing both on its left and right ends. Center-justification is commonly used for titles or headings. Figure 15.4 illustrates the use of various types of justification of text. The term hardware refers to the physical devices of a computer system. Thus, the input, storage, processing, control, and output devices are hardware. The term software refers to a collection of programs. A program is a sequence of instructions written in a language that can be understood by a computer. It is the program that controls the activity of processing by the computer and the computer performs precisely what the program wants it to do. (a) Example of left-justified (ragged-right) text.
The term hardware refers to the physical devices of a computer system. Thus, the input, storage, processing, control, and output devices are hardware. The term software refers to a collection of programs. A program is a sequence of instructions written in a language that can be understood by a computer. It is the program that controls the activity of processing by the computer and the computer performs precisely what the program wants it to do. (b) Example of right-justified text. The term hardware refers to the physical devices of a computer .system. Thus, the input, storage, processing, control, and output devices are hardware. The term software refers to a collection of programs. A program is a sequence of instructions written in a language that can be understood by a computer. It is the program that controls the activity of processing by the computer and the computer performs precisely what the program wants it to do. (c) Example of full-justified text. Introduction to Hardware and Software The term hardware refers to the physical devices of a computer system. Thus, the input, storage, processing, control, and output devices are hardware. The term software refers to a collection of programs. A program is a sequence of instructions written in a language that can be understood by a computer. It is the program that controls the activity of processing by the computer and the computer performs precisely what the program wants it to do. (d) Example of centered text heading. 5. Indenting text appropriately wherever desired such as at the beginning of a paragraph, or to set aside some portion of text from the main text in a document, or to enter text in tabular form. Tab stops are normally used for indentation. Tab stops are typically pre-defined at every fourth or fifth character, or at every Viinch from the left margin of a page layout. Most word-processing packages allow you to or at every ½-inch from the left margin of a page layout. Most wordprocessing packages allow you to change or define new tab stop positions. The Tab key on the keyboard allows you to move the cursor to the next tab stop while entering text. Figure 15.5 illustrates a few uses of indentation. A software package is a group of programs that solve a specific problem or perform a specific type of job. For example, a word-processing package may contain programs for text editing, text formatting, drawing graphics, spelling checking, etc. Thus a multipurpose computer system has several software packages, one each for every type of job it can perform. (a) Example of paragraph indentation.
A software package is a group of programs that solve a specific problem or perform a specific type of job. For example, a word-processing package may contain programs for text editing, text formatting, drawing graphics, spelling checking, etc. Thus a multipurpose computer system has several software packages, one each for every type of job it can perform. (b) Example of indenting some portion of text to set it aside from the main text. 6. Creating numbered or bulleted list of items. As shown in Figure 15.6, this feature allows descriptive items to be presented as a numbered or bulleted list for better comprehension and readability. Each item in the list has a separate number or bullet or some other symbol attached to its beginning. The numbers or symbols are automatically inserted by the word-processing software as soon as you press the Enter key. Some of the typical features found in a grammar checker are: 1. It highlights the use of double words (such as for for). 2. It highlights phrases with redundant words (such as very highest). 3. It highlights the misuse of capital letters (such as PraDeep or PradeeP). (a) Example of numbered list of items. In general, a computer's system software performs one or more of the following functions: • Supports the development of other application software. • Supports the execution of other application software. (b) Example of bulleted list of items. Entering Mathematical Symbols Several scientific and engineering documents often use mathematical symbols. A word-processing package allows you to enter not only plain text but also complex mathematical equations. Almost all word-processing packages support the use of a wide range of mathematical symbols to enable this. Figure 15.7 shows the use of a few mathematical symbols. Displaying Documents Word-processing packages allow you to display the contents of the document, you are currently working on, on the screen of the computer's monitor. With the contents displayed on the screen, the text cursor can be moved to any position in the document for editing the text. In case of a large document that cannot completely fit on the or screen, you can browse through the document by scrolling
a line at a time, a screen at a time, or a page at a time, by using suitable keys on the keyboard. Based on this feature, word-processing packages are of two types — one with WYSIWYG (pronounced "wiz-ee-wig") facility and the other one with ordinary text display facility. WYSIWYG is an acronym for "What You See Is What You Get", and word-processing packages with this facility can display a document on the screen in a form that closely resembles what the printed document will look like. These word-processing packages employ high-resolution graphics. On the other hand, word-processing packages with ordinary text display facility can display only ordinary text. Hence the embedded codes that the user has supplied for formatting the text are also displayed along with the document's contents as ordinary text. That is, the result of formatting is not visible on the screen and appears only on the printed document. However, such word-processing packages normally provide preview facility that permits you to see what the document will look like (almost, if not exactly) when it is printed. Today, most of the popular word-processing packages are of WYSIWYG type. Saving, Retrieving and Deleting Documents One of the most important features of a word-processing package is to save a document on a secondary storage and to retrieve it later at any time for reuse. At the time of saving a document for the first time, you must specify a file name for the document. The document is later retrieved for reuse by specifying its file name. You may even permanently remove the document from the secondary storage by deleting it when you feel that it is no more useful for you or when you have created and saved a new version of it. When you are editing a document, you should save it at frequent intervals so that you do not lose the updates made to it in case of a power failure. This is because any power failure wipes out all work that has not been saved. To take care of this problem, some word-processing packages offer an automatic save feature that saves the document you are currently working on after every few minutes. Printing Documents All word-processing packages support printing feature to allow users to print their documents on a printer for generating a hard copy of their documents. Some of the commonly supported printing facility in almost all modern word-processing packages and printers are: 1. Selective printing of a single page, specific pages, a range of pages, or the whole document. 2. Single-sided or double-sided printing. Single-sided printing prints only on one side of a paper, while double-sided printing prints on its both sides.
3. Portrait and landscape modes of printing. In portrait mode the printed lines are parallel to the shorter edge of the paper (like in this book), whereas in landscape mode the printed lines are parallel to the longer edge of the paper. 4. Previewing a document before printing it. This facility shrinks down the pages of your document so that an entire page or even facing pages can be displayed on the screen at once in almost the same form as it will appear when printed. This is especially useful for seeing how margin and other settings will finally appear in your document without having to print the document first. This avoids wastage of paper. Importing Text, Graphics and Images Most modern word-processing packages permit the integration of text, graphics and images. That is, in addition to entering text, you can also draw illustrative diagrams and graphs to support the textual matter in a document. While working on a document, you can even import text, graphics and images from some other document into your document. This feature often helps in saving tremendous time and effort because the user need not spend time in creating a piece of text, diagram, graph or image that can be easily obtained from somewhere else. In addition to importing, many word-processing packages also have document conversion facility that enables an imported document that was prepared using a different word-processing package, to be converted into a document compatible with the user's word-processing package. Searching and Replacing Text String All modern word-processing packages provide search facility that enables you to quickly search for all occurrences of a specific word, phrase, or groups of characters in a document. For example, you might have used the months November and December at several places in your document. If you want to ensure that the current month name appears at all places in the document, you can use the search facility to enable you to do this quickly. In addition to providing this simple search capability, word-processing packages also support search-and-replace facility. This facility enables you to perform selective or global replacement of a specific word, phrase, or groups of characters with a new set of characters. For example, suppose a document contains the name Pradip at several places. If we realize later that this is the wrong spelling of the name and it should be Pradeep instead of Pradip, we can use the search-andreplace facility to change all occurrences of Pradip to Pradeep in the document. Checking Spelling
Most modern word-processing packages come with an electronic dictionary and a spell checker to allow you to ensure that your document does not contain any misspelled word. A spell checker looks up every word in your document against the pre-stored -words in the electronic dictionary and alerts you if no match is found for a particular word in the dictionary. In such a situation, the spell checker highlights the word and waits for you action. You may choose to do either of the following: 1. You can correct the spelling. 2. If you do not know the correct spelling, you can ask the spell checker to give you a list of words with similar spelling and choose the correct word from the list. 3. If the word is spelled correctly but has been highlighted simply because it is not present in the electronic dictionary (for example, names of people and companies, abbreviations, acronyms, etc.), you can instruct the spell checker to ignore the word, or to add it to the dictionary. Unfortunately, spell checkers are not yet smart enough to check for the improper use of correctly spelled words. For example, it will not flag "form" when the word should be "from". Checking Grammar and Style Similar to a spell checker, some advanced word processing packages come with a grammar checker that enables you to correct grammatical mistakes and use of improper writing styles in your document. Some of the typical features found in a grammar checker are: 1. It highlights the use of double words (such as for for). 2. It highlights phrases with redundant words (such as very highest). 3. It highlights the misuse of capital letters (such as PraDeep or PradeeP). 4. It highlights text having subject and verb mismatches (such as she were). 5. It highlights punctuation errors (such as , .). 6. It highlights sentences written in the passive voice rather than the active voice (such as "The book was given by Mohan" is in passive voice whereas "Mohan gave the book" is in active voice). Using Thesaurus Some advanced word-processing packages come with an electronic thesaurus. A thesaurus provides a list of synonyms for a word (alternative words with similar meanings). While creating a document, if you feel that a word is being used too often or a word that you have used does not seem to be quite right, you can select the word and take help of the thesaurus for suggesting a list of viable alternatives.
As soon as you select a preferred word from the suggested list of words, the original word is automatically replaced with it. Creating Table-of-Contents and Index Some advanced word-processing packages also allow you to create table-ofcontents and index of key words for a large document such as a book, a manual, or a report. With the help of a simple command, you can automatically create for your document a table-of-contents with page numbers for chapters and various levels of headings, much like the one in the front matter of this book. Similarly, with the help of another command you can automatically create for your document an alphabetical index of key words that lists the page numbers for each occurrence of a key word, much like the one in the back matter of this book. Using Mail-Merge Often we need to send a letter to many persons. For example, a letter that is to be sent to all the participants of a workshop/seminar. The mail-merge feature found in some advanced word-processing packages allows you to create a file containing the letter and another file containing the list of names and addresses of the participants and then to merge the indicated name and address fields from the file containing the list into the proper locations in the letter as the letters are printed. Figure 15.8 shows a typical example of how to indicate the name and address fields on the letter to be substituted with actual names and addresses of the participants when the letters are printed. This feature enables printing of personalized letters for each participant on the list. & First Name & & Last Name& &Address& &City&, &Zip & Dear &First Name&,
Pradeep Sinha Pune University Pune, 411007 Dear Pradeep,
(a) A typical example of indicate the (b) This is how the name and address how to name and address the letter. get printed on the letter. fields on Figure 15.8. An example of mail-merge feature. SPREADSHEET PACKAGE What it is?
A spreadsheet package is a numeric data analysis tool that allows us to create a kind of computerized ledger. A manual ledger is a book having rows and columns that accountants use for keeping a record of financial transactions and for preparing financial statements. Accountants use the manual ledger with pencil, erasure and hand calculator to prepare financial statements. This is a tedious task and often takes a long time due to several iterations of formula calculations to come out with an acceptable and satisfactory financial statement. A spreadsheet package offers considerable ease of performing such tasks by automating all arithmetic calculations and making it much easier to change certain numeric values and immediately seeing the effect of these changes across the worksheet (ledger). With spreadsheet software in place, we are no longer confined to using pencils, erasers, and hand calculators for dealing with any task that requires numeric data analysis. Whereas paper ledgers were tools for accountants, spreadsheet packages are tools for anyone who needs to record, organize, or analyze numbers as rows and columns of data. That is, a spreadsheet package is useful for any numerical analysis problem whose data can be organized as rows and columns. Some typical uses of spreadsheets packages are: 1. For maintaining and analyzing inventory, payroll, and other accounting records by accountants. 2. For preparing budgets and bid comparisons by business analysts. 3. For recording grades of students and carrying out various types of analysis of the grades by educator 4. For analyzing experimental results by scientists and researchers. 5. For tracking stocks and keeping records of investor accounts by stockbrokers. 6. For creating and tracking personal budgets, loan payments, etc. by individuals. This list is by no means exhaustive and is given here only to give an idea of how flexible and versatile tool is spreadsheet package. Commonly Supported Features An example of a sample spreadsheet is shown in Figure 15.9. Some of the key features of spreadsheet package are explained below with the help of this example. Rows and Columns A spreadsheet is organized in a tabular form with rows and columns. The rows are identified by numbers, whereas the columns are identified by letters. Single letters are used to identify the first 26 columns and double letters are used for subsequent columns (A, B, ..., Z; AA, AB, ..., AZ; BA, BB, ..., BZ; ...). A large spreadsheet may have hundreds of rows and hundreds of columns.
Cells The intersection of a row and a column is called a cell. A cell is identified by its address, which is a combination of its column letter and row number (such as F4, C18). Data are stored in cells. Range of Cells Many spreadsheet operations involve a range of cells instead of a single cell. For instance, in Figure 15.9, the contents of cell E5 is equal to the sum of the contents of cells B5, C5 and D5, which is a range of three cells in a row. A range of cells may involve many cells and entering every individual cell for an operation to be performed on them may be cumbersome and time taking. To take care of this problem, all spreadsheet packages allow the use of following types of cell ranges: 1. Row Range. It is used to include a set of adjacent cells in the same row. For example, the range B5..D5 refers to the cells B5, C5 and D5. 2. Column Range. It is used to include a set of adjacent cells in the same column. For example, the range C5..C9 refers to the cells C5. C6. C7. C8 and C9. 3. Block Range. It is used to include a set of adjacent cells in a rectangular group of cells. For example, the range B5..D9 refers to the cells B5, B6, B7, B8, B9, C5, C6, C7, C8, C9, D5, D6, D7, D8 and D9. Notice that a particular range is indicated by the addresses of the endpoint cells separated by two periods (B5..D9) oracolon(B5:D9). Cell Contents Different types of data can be entered into a cell. The four commonly used types of cell data are: 1. Label. A label is any string of alphanumeric text. Labels are used either to enter alphanumeric values in cells (such as in cells A5 to A9 of Figure 15.9) or to enter descriptive data or titles to make the contents easier to understand (such as in the cells in rows 1 and 3 of Figure 15.9). Notice from the figure that the label in cell Bl extends across columns C, D, E and F. This is possible when the adjacent cells are blank, otherwise the text gets truncated after the column width.
2. Numeric Value. A numeric value is a number on which the spreadsheet can perform calculations. For example, in the spreadsheet of Figure 15.9, cells B5..D9 contain numeric values. 3. Formula. A formula may include numbers, mathematical operators (such as +, -, *, /, etc.), mathematical functions supported by the spreadsheet (such as SUM, AVG, MIN, MAX, etc.), and cell addresses. For example, cell E9 contains the formula @SUM(B9..D9) which means that the sum of the numeric values of cells B9, C9 and D9 should be displayed in cell E9. The same result could have also been obtained by entering the formula B9+C9+D9 in cell E9. Cell F9 of the spreadsheet of Figure 15.9 also contains a formula, which is E9/3 and calculates the percentage marks obtained by the student named J. Smith. Notice that in case of formula entry, the contents displayed in the cell is not the formula itself but the result of formula calculation. 4. Date and Time. Although not used in the spreadsheet of Figure 15.9, the date and/or time contents type is normally used in a spreadsheet to display the actual date and time whenever the spreadsheet is opened for use, or if calculations are to be made based on date or time. For example, in case of a spreadsheet used for interest calculation, it is required to figure out the number of days between two dates. Cell entries with data type as date are helpful for such applications. Relative and Absolute Cell Addresses When a formula entered in a cell has references to other cells (contains cell addresses), by default, the references are relative to the cell with the formula in it. For example, cell E5 contains the formula @SUM(B5..D5) that adds the contents of cells B5, C5 and D5. This formula can be entered once in cell E5 and then copied in cells E6, E7, E8 and E9. Because the cell address is relative, this copying will result in cell E6 having the formula @SUM(B6..D6), cell E7 having the formula @SUM(B7..D7), and so on. Similarly, the formula +E5/3 can be entered once in cell F5 and then copied in cells F6, F7, F8 and F9. Notice that the facility of relative cell addressing can be of great help in creating a spreadsheet quickly by entering the formulas once and copying them to other cells as in the example of Figure 15.9. If a formula requires a cell address, which should not change even when the formula is copied to other cells, you can use absolute cell addressing to accomplish this. Depending on the requirement, a cell reference can be made fully or partially absolute as follows: 1. Precede both the column letter and the row number of the cell with a dollar sign ($) to make both column and row references absolute. For example, $A$2 is a reference to cell A2 that is absolute as to both row and column. When copied at any other cell location, it will always reference the cell A2.
2. Precede only the column letter of the cell with a dollar sign ($) to make only the column reference absolute. For example, $A2 is a cell reference that is absolute as to column only. When copied at any other cell location, it will reference the cell in column A of the current row. That is, if a formula entered in cell A2 containing $A2 is copied in cell El2, this cell reference in the formula in cell E12 will refer to cell A12. 3. Precede only the row number of the cell with a dollar sign ($) to make only the row reference absolute. For example, A$2 is a cell reference that is absolute as to row only. When copied at any other cell location, it will reference the cell in row 2 of the current column. That is, if a formula entered in cell A2 containing A$2 is copied in cell El2, this cell reference in the formula in cell E12 will refer to cell E2. Spreadsheet Commands All spreadsheet packages come with a set of commands to enable you to perform many different operations. For example, copy and move commands enable you to copy/move the contents of one or more cells to some other cells. Insert and delete commands enable you to add/remove a column or row anywhere in your spreadsheet. Format commands enable you to control the way cell contents are displayed. Formatting features include specifying the width of columns, the number of decimal places in a non-integer numerical value, automatic insertion of a dollar sign and commas to separate thousands for currency amounts (such as $6,852), the use of right- or left-justified values or labels, etc. Save and retrieve commands enable you to store the format and data of an already created spreadsheet on a disk, and to retrieve it later for updates and analyses. Print command enables you to generate a hard copy of all or part of a spreadsheet. Spreadsheet Graphics Most spreadsheet packages come with presentation graphics feature, which enables you to create graphs and charts from numerical data stored in a spreadsheet. As shown in Figure 15.10, the most popular types of graphs and charts supported for graphical representation of numerical data are line graphs, bar charts and pie charts. This feature is very useful in presenting spreadsheet data in easier to understand visual forms because most people find it difficult to interpret a table of numbers. GRAPHICS PAKCAGE What it is?
Graphics packages enable us to use a computer system for creating, editing, viewing, storing, retrieving and printing designs, drawings, pictures, graphs and anything else that can be drawn in the traditional manner. Commonly Supported Features Graphics packages normally support the following features: 1. Draw designs 2. Paint drawings and pictures 3. Present graphs 4. Drag-and-drop graphic objects 5. Import graphic objects 6. Screen capture Each of these features may be provided as a separate graphics package (such as design package, paint package, presentation package, etc.), or they may be integrated together in a single graphics package, or they may be included in some other application package (for example, a word-processing package may have a design package included in it to enable users to draw figures in their documents, or a spreadsheet package may have a presentation package included in it to enable users to present numerical data in the form of graphs). No matter how they are supported, the features remain the same and our main interest here is to know about these features rather than how they are supported. What each of these features enables us to do is briefly described below. Draw Designs This feature enables users to draw graphics objects, such as lines, circles, rectangles, arcs, etc., to create diagrams and designs. Users need not worry about drawing straight lines or exact circles. The system automatically makes the lines, circles, arcs, etc. smooth and properly connected to each other. The system also allows users to move, copy, delete, rotate, tilt, flip horizontally or vertically, increase or decrease the size of the graphic objects. With all these facilities, users can draw complex designs with great ease. In fact, computer-aided design (CAD) is an area that is mainly based on this feature of graphics software, deals with the integration of computers and graphics design packages for the purpose of automating the design and drafting process. CAD systems are used by architects and engineers to create architectural drawings, product designs, landscaping plans, and many different types of engineering drawings. They enable designers to work much faster than they once worked in the era of manual drafting, creating in few hours designs that used to take several days.
Paint Drawings and Pictures This feature enables users to create and modify drawings and pictures in the form of images. Unlike the draw feature (discussed above), which uses vector graphics for composing graphic objects, the paint feature uses raster graphics for composing graphic images. In vector graphics, the design is composed of patterns of lines, points, circles, arcs, and other geometric shapes (vectors) which can be easily represented by few geometric parameters (for example, a line is represented by recording the x and y coordinates of two points, and a circle is represented by recording the x and y coordinates of a point and the radius of the circle). On the other hand, in raster graphics, the image is composed of patterns of dots called pixels (picture elements). That is, a painting software ere image by turning individual screen pixels on or off. Because the image is mapped on to the screen based on binary bits, this technique is called bit mapping and an image represented in this manner is called a bit-mapped image. The number of bits needed to describe a pixel increases with the monitor's resolution and the number of colors that can be presented. Obviously, vector graphics takes up much less storage space than raster graphics for the same drawing. However, raster graphics for provides more flexibility in drawing complex shapes and representation of colors. That is, painting software give the users a greater range of creativity than drawing software provide. Present Graphs This feature enables users to create graphs and charts from numerical data. The numerical data to be converted into a graph or chart may be imported from another software, such as a spreadsheet or a database. Among the most popular types of graphs and charts used for graphical representation of numerical data are line graphs, bar charts and pie charts (see Figure 15.10). Line graphs contain one or more lines connecting data points plotted on a horizontal and vertical axes. Bar charts use one or more bars on a horizontal and vertical axes to show values by the lengths of the bars. Sometimes^ bar is^divided into component parts; and the sum of all the parts equals the total length of the stacked bar. Pie charts use one or more circles divided into sectors to show the component parts of a whole. Single circles can be used, or several can be drawn to compare changes in the component parts over time. Regardless of the type of representation used, a user can vary colors, add and delete lines and headings, change scales and edit the graphs or charts suitably to get the desired results. Some presentation graphics software also have animation capability that enables users to make the graph or chart dynamically change on the screen to depict increase or decrease of figures over time. Presentation graphics software is very useful for analysts and decision makers because it allows them to gain a better understanding of the relationships, changes,
and trends that are buried in their numeric data. This helps them in making more informed decisions. Drag-and-drop Objects This feature enables the users to create their overall designs and pictures much faster by allowing them to use ready-made graphic objects or images, supplied with the software. For example, the drawing software comes with a set of readymade shapes like line, rectangle, circle, etc. which the user can use in his/her design as per the requirement. The user can select a desired object from the set of given objects and drag it to the desired position on the drawing area and then drop it there to add it to the overall design. The drag feature also allows the user to suitably change the size or the shape of the object to match with the overall design. It also enables the user to glue multiple objects together such that they stay connected and move together if one of them is moved on the drawing area. Similarly, the painting software comes with a clip art library of stored images. The user can select a desired image from the clip art library and drag it to the desired position on the drawing area and then drop it there to add it to the overall picture being created. The user can also change the size, color, texture, etc. of the image to match with the overall picture. With this feature, the user need not spend time in creating an image that is readily available in the clip art library. Import Objects This feature further enables the users to create their overall designs and pictures much faster by allowing them to use not only the graphic objects or images supplied with the software, but even those graphic objects or images that are stored in files created by some other software or means. For example, a photograph may be scanned and the scanned image may be imported into and included in an overall picture that the user is currently creating. Similarly, a graphics object or image received from someone through electronic mail can be imported in a drawing from the mail file. With this feature, the user need not spend time in creating an image or an object that can be readily obtained from somewhere else. Screen Capture This feature enables the users to take a snapshot of a screen display and to convert it into an image that can be stored in a file and later imported into a document. This feature is very useful while writing books or manuals that need to include screen images as illustrations. A screen is captured as a bit-mapped image. Once stored in a file, it can be imported into any document at any time and even manipulated with a paint program.
PERSONAL ASSISTANCE PACKAGE What it is? Personal assistance packages allow individuals to use personal computers for storing and retrieving their personal information, and planning and managing their schedules, contacts, finances and inventory of important items. Commonly Supported Features Features normally supported by personal assistance packages are as follows (some may support only a few of these, while others may have several additional features not listed here): 1. Calendar 2. To-do list 3. Address book 4. Investments book 5. Inventory book What each of these features enables us to do is briefly described below. Calendar This feature enables users to record their appointments and plan their schedules on a yearly, monthly, weekly, daily and even hourly basis. Unlike a hard-copy calendar which can also be used by individuals to record appointments and plan schedules, a calendar software is designed to automatically alert a user prior to a scheduled meeting with an audible alarm or a pop-up message on the display, warn the user of possible schedule conflicts when a new appointment is recorded, and automatically schedule tasks of recurring nature. It also allows users to view or print the highlights of past or future events at the desired level of granularity (yearly, monthly, weekly, hourly, etc.). To-do List This feature enables users to record, prioritize and manage the things to do. A user can dynamically keep entering the tasks to do as and when they come up, and at any point of time the user can instruct the system to display or print the tasks by project, deadline date, or priority. The software also allows users to add or delete tasks to update the to-do list, or to include additional information against a task to
record its current status or any other information related to the execution of the task. Address Book This feature enables users to record the names, addresses, telephone numbers, affiliations, date of births, etc. of their friends, relatives, clients, and other contact persons. The entries in the address book can be made as and their friends, relatives, clients, and other contact persons. The entries in the address book can be made as and when they come up, and at any point of time the user can ask the system to provide details of a particular person, or to display or print a list of persons having something in common. For example, from the address book, a user can easily obtain a list of persons living in a particular city, persons whose date of birth falls on a particular date, persons affiliated to a particular organization, etc. Investments Book This feature enables users to maintain a record of their various types of investments or loans such as long-term and short-term deposits in banks, stocks, bonds, housing loan, car loan, life insurance, medical insurance, etc. The software automatically alerts a user prior to the maturity of a term deposit, or prior to the due date of the premium payment for a loan or insurance with an audible alarm or a pop-up message on the display. It also allows users to view or print the summary or details of their investments or liabilities at any instance of time. The software can also be fed with the information of an individual's daily, weekly, or monthly income and expenditure statements which when clubbed with the investments and liabilities information, helps in effective financial planning by the individual. Inventory Book This feature enables users to maintain a record of the goods in their houses along with associated information for each item, such as when was it purchased, from where was it purchased, what is the duration of its warranty period, is it insured or not, etc. In addition to making the list of all available items in a house readily available, this feature also enables the user to find out the details of an item when there is a break down in it or when it is stolen. Points to Remember 1. A word-processing package enables us to create, edit, view, format, store, retrieve and print documents on a computer system. 2. Some of the commonly supported features found in modern word-processing packages are entering text; editing text; formatting page style; formatting text;
entering mathematical symbols; displaying documents; saving, retrieving and deleting documents; printing documents; importing text, graphics and images; searching and replacing text string; checking spelling; checking grammar and style; using thesaurus; creating table-of-contents and index; and using mail-merge. 3. A style sheet is a pre-stored page format that can be used while creating a new document or can be applied to an existing document. 4. A font is a complete set of characters with the same style and size. A wordprocessing package comes with several standard fonts. 5. Font size is measured in points. A point is 1/72 of an inch, and the size refers to the distance from the top of the tallest character to the bottom of the character that extends the lowest. 6. The three commonly used font styles are italic, bold and underline. 7. Justification is the alignment of text on the left or the right margin, or on both margins. The four types of justification are left-justification, right-justification, full-justification and center-justification. 8. A word-processing package with WYSIWYG feature can display a document on a screen in a form that closely resembles what the printed document will look like. 9. The preview feature of a word-processing package shrinks down the pages of your document so that an entire page or even facing pages can be displayed on the screen at once in almost the same form as it will appear when printed. 10. The search-and-replace feature of a word-processing package enables you to perform selective or global replacement of a specific word, phrase, or groups of characters with a new set of characters. 11. The spell-checker of a word-processing package helps you to ensure that your document does not contain any mis-spelled word. 12. The grammar-checker of a word-processing package enables you to correct grammatical mistakes and use of improper writing styles in your document. 13. A spreadsheet package is a numeric data analysis tool that allows us to create a kind of computerized ledger. It is useful for any numerical analysis problem whose data can be organized as rows and columns. 14. Some of the key features of modern spreadsheet packages are support for a large number of cells, support for addressing a range of cells by the addresses of the endpoint cells, support for different types of cell data (such as, label, numeric value, formula, and date and time), support for use of relative and absolute cell addresses in formula, support for a wide range of commands, and support for displaying numeric data in the form of graphs and charts. 15. A graphics package enables us to use a computer system for creating, editing, viewing, storing, retrieving and printing designs, drawings, pictures graphs and anything else that can be drawn in the traditional manner. 16. Some of the commonly supported features found in modern graphics packages are drawing designs, painting drawings and pictures, presenting graphs and charts,
dragging-and-dropping graphic objects, importing graphic objects, and capturing screen snapshots. 17. Computer-aided-design (CAD) deals with the integration of computers and graphics design packages for the purpose of automating the design and drafting process. 18. The two methods commonly used for composing graphic objects are vector graphics and raster graphics. In vector graphics, the graphic object is composed of patterns of lines, points, circles, arcs and other geometric shapes which can be easily represented by few geometric parameters. On the other hand, in raster graphics, the graphic object is composed of patterns of dots called pixels. Vector graphics representation takes up much less storage space than raster graphics representation for the same graphic object. However, raster graphics provides more flexibility in drawing complex shapes and representation of colors. 19. A. personal-assistance package allows individuals to use personal computers for storing and retrieving their personal information, and planning and managing their schedules, contacts, finances and inventory of important items. 20. Some of the commonly supported features found in modern personalassistance packages are calendar, to-do-list, address book, investments book and inventory book. Questions 1. What is a word-processing package? List out some of the key features supported by modern word-processing packages. 2. Differentiate between the insert and typeover modes of text entry in a wordprocessing package. 3. Differentiate between the cut-and-paste and copy-and-paste facilities of a word-processing package. 4. What is a style sheet in a word-processing application? How is it useful? 5. What is a font? What is meant by font size and font style? 6. Differentiate between left- and right-justified text. 7. What is meant by WYSIWYG facility? How is it useful? 8. What is print-preview facility? How is it useful? 9. Differentiate between portrait and landscape modes of printing. 10. What is meant by text, graphics and image importing facility? How is it useful? 11. Explain the search-and-replace facility of a word-processing package. 12. What is a spell checker in a word-processing package? How does it work? 13. What is a grammar checker in a word-processing package? List out some of its typical features found in a modern word-processing package. 14. How does the presence of a thesaurus in a word-processing package help? 15. What is mail-merge facility? How is it useful?
16. What is a clip art library? Describe how it helps in document preparation with a word processor. 17. List out any five features of a word-processing package. Explain how each of these five features is helpful in document preparation. 18. What is meant by word-wrap feature? How is it useful? 19. Give an example of a situation when the global search-and-replace command of a word-processing package will prove useful. 20. What options does a spell checker typically present to the user when it encounters an unidentified word in a document? 21. A spell checker cannot check for the improper use of correctly spelled words. Explain why? 22. What is a spreadsheet package? List out some of its typical uses. 23. Think of an application for which a spreadsheet package is useful. Now draw a sample spreadsheet for 1 application. 24. What is a cell in a spreadsheet? How is a single cell identified? How is a range of cells identified? 25. Differentiate among row range, column range and block range of cells in a spreadsheet. 26. List out the four commonly used types of cell data found in a spreadsheet along with the need for each type. 27. Differentiate between relative and absolute cell addresses in a spreadsheet. Explain the need for these two types of cell addresses. 28. Explain how a cell address can be made fully or partially absolute when used in a formula. 29. List out some of the formatting features supported in a modern spreadsheet package. 30. Explain about the graphics feature of modern spreadsheet packages. How is it useful? 31. List out five commands that are likely to be present in a spreadsheet package. Explain the use of each command. 32. What is a graphics software? List out some of the features normally supported in a graphics software. 33. What is the major difference between the paint and the draw features of a graphics software? 34. What is a CAD system? List out some of its typical uses. 35. Differentiate between vector graphics and raster graphics. Give their relative advantages and disadvantages. 36. What is a bit-mapped image? Why is it so called? 37. Explain how the following features of a graphics software are useful: (a) Present graphs (c) Import objects (b) Drag-and-drop objects (d) Screen capture 38. What is a personal assistance package? List out some of the features normally supported by it. Explain the usefulness of each of these features.
39. Write short notes on: (a) Spreadsheet package (b) Graphics package 40. Write short notes on: (a) Computer Aided Design (CAD) (b) Vector graphics (c) Raster graphics (c) Word-processing package (d) Personal assistance package 41. Write short notes on the following features of a word-processing package: (a) Word-wrap (e) Page formatting (b) Text editing . (f) Text formatting (c) WYSIWYG (g) Search-and-replace (d) Print preview (h) Import text, graphics and image 42. Write short notes on the following facilities provided by some advanced word-processing packages: (a) Spell checking (b) Grammar and style checking (c) Mail-merge 43. Write short notes on the following with reference to a spreadsheet package: (a) Cell contents (c) Graphics feature (b) Range of cells (d) Relative and absolute cell addresses
Chapter 16 Business Data Processing This chapter deals with the basic concepts of business data processing. In this chapter, you will learn about the following: 1. Difference between data and information and how data processing converts raw data into useful information. 2. The data storage hierarchy commonly used to facilitate data processing. 3. The standard methods of organizing data. 4. The basic concepts of database systems. 5. The modern concepts in database technology. WHAT IS DATA PROCESSING? Data is a collection of facts - unorganized but able to be organized into useful information. A collection of sales orders, employee time sheets, and class attendance cards are a few examples. Data can be manipulated to produce output, such as bills, employee salary slips, and student attendance reports. This output, which can be used to help people make decisions, is called information. That is, information is data arranged in an order and form that is useful to the people who receive it. Processing, in general terms, is a series of actions or operations that converts some input into useful output. When we speak of data processing, the input is data, and the output is useful information. Hence, data processing is defined as a series of actions or operations that converts data into useful information. It consists of three sub-activities - capturing the input data, manipulating the data, and producing the output results. A data processing system includes the resources such as people, procedures, and devices that are used to accomplish the processing of data for producing desirable output. Thus, data is the raw material of information and just as raw materials are transformed into finished products by a manufacturing process, raw data is transformed into information by data processing. DATA STORAGE HIERARCHY In data processing, storage of data is often conceptualized as a data storage hierarchy, which normally consists of the following six levels (see Figure 16.1). Level 0 - Bit - A single binary digit (0 or 1).
Level 1 - Character -Multiple related bits are combined to form a character (byte). Level 2 - Field - Multiple related characters are combined to form a field. Level 3 - Record - Multiple related fields are combined to form a record. Level 4 - File - Multiple related records are combined to form a file. Level 5 - Database - Multiple related records are combined to form a file. 1. Bit. The smallest item of data is a single binary digit (a bit), either a 0 or a 1. 2. Character. Multiple related bits are combined to form a character (or byte). For example, the letter M, the number 8, and the special character $ are characters. Whereas the bit is the basic unit of primary and secondary storage, the character is the basic unit for human perception. 3. Field. Multiple related characters are combined to form a field. For example, if we are processing employees' data of a company, we may have an employeecode field, an employee-name field, hours worked field, an hourly-pay-rate field, a tax-rate-deduction field, etc. A field is a meaningful collection of related characters. It is the smallest logical data entity that is treated as a single unit in data processing. Note that a field may sometimes have a single character, such as sex code may be represented as M or F. 4. Record. Multiple related fields are combined to form a record. For example, an employee record will contain the fields containing the data of an employee, such as the employee's code, name, hours worked, pay-rate, tax-rate deduction, and so forth. Similarly, a student record will contain the fields containing the data of a student. Thus, a record is a collection of related fields that are treated as a single unit. 5. File. Multiple related records are combined to form a file. For example, a collection of all the employee records of a company would be an employee file. Similarly, a collection of all the inventory records of a company forms an inventory file. Thus, a. file is a number of related records that are treated as a unit. Notice that every record in a file has the same set of fields. Each record in a file is identified, for storage and retrieval purposes, by a key field whose contents are unique for each record in a file. In case of our employee file, employee-code field may serve as the key field. 6. Database. Multiple related files are integrated to form a database. For example, an employee database of an organization may integrate the records of multiple employee files, such as employee salary file containing details of salary and taxation information of all employees, employee personnel information file containing details of personnel information of all employees, and employee skillset file containing details of various types of skills of all employees. In a database, the data from multiple files are integrated in such a manner so that data redundancy is minimized. For example, in case of the above example of three different employee files, if in each file an employee record contains an employee address field, and if an employee changes his/her residence, the address must be changed in all the three files. In a database, the three files will be integrated in
such a manner that an employee's address data is stored only once and made available to all applications. Therefore, only one update will be needed in case of change of address of an employee. In essence, a database is a collection of logically related data elements from multiple files. STANDARD METHODS OF ORGANIZING DATA The two standard methods used in data processing for organizing data are 1. File-oriented approach and 2. Database-oriented approach. The file-oriented approach was the traditional method used in early days of data processing and has largely been replaced today by the database-oriented approach. However, several applications, dealing with simple and small data sets, use the file-oriented approach even today. Hence we will have a look at both these methods of organizing data. File-oriented Approach In this method, an application's data is organized into one or more files and the application program processes the data stored in these files to generate the desired output. For example, it is customary to set up a master file of permanent (and, usually, the latest) data, and to use transaction files containing data of a temporary nature. For instance, in a payroll processing application, the master payroll file will contain not only all the permanent details about each employee, his/her name, address, employee code, pay-rate, income tax rate, and so forth, but also the current gross-pay-to-date total and the tax paid-to-date total. The transaction payroll file will contain details of hours worked this week, normal and overtime, and if piecework is involved, the quantity of goods made. When the payroll program is processed, both files will have to be consulted to generate this week's pay-slips, and the master file updated in readiness for the following week. In the file-oriented approach of organizing data, a set of programs is provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their files. All these programs together form a File Management System (FMS). We will have a look at the commonly supported features in a typical file management system in the next section of this chapter. Database-oriented Approach The file-oriented method of organizing data for data processing applications is simple, inexpensive, and usually easy to use. However, it suffers from the following limitations:
1. Limited query flexibility. In file-oriented approach, the records in a file are organized according to a key field. As long as we search records based on key field value, the query can be processed quickly. However, when the key field is not relevant to the information needed, the entire file has to be searched. For example, if the key field is employee code in an employee file, and if we need to list out the names of all employees with a certain educational background, a search of all the file records will be required to produce this information. If this information is frequently needed, a new file structured on educational background key will have to be created along with a new program to process this file. 2. Data redundancy. In file-oriented approach, the same data items are often included in many different files. For instance, in the above example of creating a new file structured on educational background key, the new file will also have several of the data items stored in the first employee file. Similarly, suppose a bank uses the file-oriented approach for storing its customers data, and it maintains separate customer files for its various types of services such as savings accounts, term deposits, and loan accounts. In this case, a great deal of redundant data for a bank customer (such as home address, age, occupation, and identification information) may be contained in multiple files. Repetition of the same data items in more than one file is known as data redundancy. It leads to increase in the cost of data entry and data storage. It also leads to data integrity problem described below. 3. Data integrity problem. Data integrity is the consistency of the data in all files. That is, in file-oriented approach, when some change occurs in a data item, every file, which contains that field, should be updated to reflect the change for consistency. For instance, in our example in which a bank maintains separate customer files for each type of account, when a customer moves to a new address, his/her address field must be updated in all the customer files in which this customer's record is present. Integrity of data is necessary to avoid confusions that may result when one file is updated while others are not. For example, in case of a change in a data item, if all the files containing the data item are not updated correctly, there may be frequent discrepancies among reports produced fro: different files, causing confusion. 4. Lack of "program/data independence. In file-oriented approach, the application programs usually contain data format statements that precisely define each data field to be processed. This often results in different files having the same data item stored using different data formats. Data dependence occurs when, the data is dependent on the application. Due to data dependence problem, whenever there is a need to add, delete, or change data formats, the application program must also be changed. Data dependence problem also causes incompatibility among data files
from different applications due to which these data files cannot be linked if such a need arises. 5. Limited data security flexibility. A file-oriented approach normally offers filelevel data security feature. That is, data access restrictions can be enforced only for an entire file and not for a record or a field of data item. Dissatisfied with these limitations of the file-oriented approach of organizing data, researchers began looking for a better method of organizing data to consolidate activities. Their efforts resulted in the database-oriented approach for organizing data. In this approach, data from multiple related files is integrated together in the form of a database, which has the following properties: 1. It provides greater query flexibility. 2. It reduces data redundancy. 3. It solves data integrity (inconsistency) problem. 4. It makes data independent of the application programs. That is, fields can be added, changed, and deleted from the database without affecting existing programs. 5. It also includes data security features at database level, record level, and even at field level to provide greater flexibility of restricted data access. In the database-oriented approach of organizing data, a set of programs is provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their data in a database. All these programs together form a Database Management System (DBMS). Later in this chapter, we will have a look at the commonly supported features in a typical database management system. FILE MANAGEMENT SYSTEM In the file-oriented approach of organizing data, a set of programs is provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their files. All these programs together form a File Management, System (FMS). The commonly supported features in a typical file management system are described below. File Types Data files are categorized according to the way the application uses them. A file management system typically supports the following types of files: 1. Transaction File. A transaction file is used to store input data until it can be processed. For example, in a payroll processing application for generating payslips of employees on a weekly basis, the transaction file will contain weekly data of each employee such as details of hours worked this week, normal and overtime, and if piecework is involved, the quantity of goods made.
2. Master File. A master file contains all the current data relevant to an application. For example, in the payroll processing application mentioned above, the master file will contain the permanent details about each employee (his/her name, address, employee code, pay-rate, income tax rate, and so forth), and also the current gross-pay-to-date total and the tax paid-to-date total. When the payroll program is processed, both the master file and the current week's transaction file are consolidated to generate this week's pay-slips, and the master file is updated in readiness for the following week. 3. Output File. Some applications use multiple programs for data processing. In such applications, often the output produced by one program is used as input to another program. Hence the produced output by the former program is stored in a file known as output file, which is later on used by the latter program. 4. Report File. A report file holds a copy of a report generated by a data processing application in computer-accessible form until it is convenient to print it. It is often advantageous to keep report files instead of paper documents because the files are easier to store and carry. 5. Backup File. A backup file is a copy of a file, created as a safety precaution against loss of data caused due to corruption or inadvertent deletion of the original file. Regular creation of backup files is extremely important. File Organizations File organization deals with the physical organization of the records of a file for the convenience of storage and retrieval of data records. System designers choose to organize, access, and process the records of various files in different ways depending on the type of application and needs of the users. The three commonly used file organizations in business data processing applications are 1. 2. 3.
Sequential, Direct/Random, and Indexed Sequential
The selection of a particular file organization depends upon the type of application. The best organization to use in a given application is the one that meets the application's data access and processing requirements in the most effective and economical manner. Hence an application designer must evaluate the strengths and weaknesses of each file organization before making a choice of the type of file organization to be used for the application at hand. File organization requires the use of some key field or unique identifying value that is found in every record in the file. The key value must be unique for each record of the file because duplications would cause serious problems. In the payroll example, the employee code field may be used as the key field.
Sequential Files In a sequential file, records are stored one after another in an ascending or descending order determined by the value of the key field of the records. In payroll example, the records of the employee file may be organized sequentially by employee code sequence. The principal storage medium for sequential files is magnetic tape. The computer processes a sequential file in the order in which data is stored in the file. That is, the first record is read and processed first, then the second record is read and processed, and so on. For instance, to locate a particular record in a sequential file, the computer reads each record in sequence from the beginning of the file and compares its key field value to the one that is needed. The search ends only when the desired key value matches with the key field value of the currently read record. Notice that with this processing mechanism, if only a single record has to be processed, on an average, about half the file has to be searched to retrieve the desired record for processing. Thus, sequential files are not suitable for applications that process only one or a few records at a time. Sequential file organization is the most efficient and economical file organization in case of applications in which there is a large number of file records to be updated at regularly scheduled intervals. That is, when the activity ratio (the ratio of the total number of records in transaction file and the total number of records in master file) is very high. This can be ensured by using the idea of batch processing, in which transactions of a similar type are accumulated into batches, then these batches are sorted into sequence with respect to the key field values of the records, and then the entire batch of records is processed in a single pass through the file. Applications such as payroll processing and monthly bill processing are processed in this manner. Advantages of Sequential Files Sequential files typically have the following advantages: 1. They are conceptually simple to understand and use. 2. They are also easy to organize and maintain. 3. They need relatively inexpensive I/O media and devices for their storage and processing. 4. They are most efficient and economical to use in applications in which the activity ratio is high (me records are changed during update runs). Disadvantages of Sequential Files
Sequential files typically have the following disadvantages: 1. They prove to be very inefficient and uneconomical for applications in which the activity ration is very low. 2. Since an entire sequential file may need to be read just to retrieve and update few records, accumulation of transactions into batches is recommended before processing them. Hence the use of sequential files is limited to a batch processing environment. 3. Because of the need to accumulate transactions into batches before processing them, sequential processing precludes the possibility of up-to-the-minute data. 4. Sequential processing requires the files to be sorted before processing. That is, both the transaction and master files must be sorted and placed in the same sequence prior to processing. 5. Sequential processing often leads to data redundancy problem, since the same data may be stored in several files sequenced on different .keys. Direct Files Many applications require up-to-the-minute timeliness of data. The users of such applications cannot afford to wait for the transactions to be accumulated in batches and then processed together. Such applications require a transaction to be processed immediately as and when it occurs. A few examples of such applications are airline or railway reservation systems, teller facility in banking applications, systems for enquiring whether a certain item is in stock in a store, etc. Note that the activity ratio is very low in these applications since only one transaction is processed at a time. Hence the use of sequential files for such applications proves to be very inefficient and uneconomical. The use of direct/random files is recommended for such applications because in a direct/random file organization, the desired record pertaining to the current transaction at hand can be directly located by its key field value without having to search through a sequence of other records. For example, when a passenger makes a reservation request for a particular train on a particular date, the computer can directly access the record for the train and update its seats position for that date. Obviously, direct/random files need to be stored on a direct access storage device to enable direct access of records. The principal storage medium for direct/random files is magnetic/optical disk. The obvious question that arises now is how are the records in a direct file physically organized so that it becomes possible to directly access a particular record, given its key field value, from a number of records stored in the file. To enable this, direct file organization uses an address generating function to convert a record key value into a storage address on the disk on which the file is stored. Each record in the file is stored at the location to which the address generating
function maps the record's key field value. The process is known as hashing and the address generating function is called a hashing algorithm. Although the hashing algorithm is selected in such a manner so as to generate a unique address for each record key value, but in practice, the hashing algorithm sometimes maps the key values of two or more records to the same storage address. This problem is known as collision. When a collision occurs, it may be taken care of in many ways. One method is to use a pointer at the address calculated by the hashing algorithm. This pointer points to a linked list of addresses, which contains the physical addresses of all records whose key values map to the address calculated by the hashing function. To search a record given its key value, the computer applies the hashing algorithm on the given key to generate its corresponding address. If the record is found at the calculated address, the search is over. Otherwise, the computer traverses the linked list pointed to by the pointer stored at the calculated address, accesses the addresses stored in the linked list and the corresponding records one-by-one, and compares the given key with the key of the accessed record. The search process finishes as soon as the record with the matching key is found. If required, the records of a direct file may also be processed sequentially in ascending or descending sequence of key field value. However, if a large number of records need to be processed sequentially (that is, if the activity ratio is high), direct files prove to be very inefficient as compared to sequential files because the computer has to use the hashing algorithm and reposition the read/write head of the disk for every record that has to be processed. Advantages of Direct Files Direct files typically have the following advantages: 1. Given the key, any record can be quickly located and retrieved directly without the need for sequential search of the file. 2. Transactions need not be sorted and placed in sequence prior to processing. 3. Accumulation of transactions into batches is not required before processing them. They may t processed as and when generated. 4. It can support interactive online applications that need to provide up-to-theminute information in response to inquiries from users. 5. If required, it is also possible to process direct file records sequentially in a record key sequence. Disadvantages of Direct Files Direct files typically have the following disadvantages:
1. They require relatively expensive hardware and software resources because they must be stored on direct-access storage device such as a disk. 2. Due to address generation overhead involved, they are less efficient and economical than sequential files for use in sequential applications with high activity ratio. 3. Special security measures are often necessary for online direct files that are simultaneously accessible from multiple online stations. Indexed Sequential Files We are all familiar with the concept of an index. For example, the display board containing the names and room numbers of the various occupants at the entrance lobby of a large multistoried building is an index that helps us locate a particular person's room within the building. For instance, to find the room of Dr. Sharma within the building, we would look up his name in the directory (index) and read the corresponding floor number and room number. This idea of scanning a logically sequenced table is preferable to searching door-by-door for the particular, name. Similarly, if we wished to read the section in this book about printers, we would not begin on page 1 and read every page until we came across the topic of interest. Rather, we would find the subject in the contents (which serves as an index) or the index at the back of the book to locate the page number, and then turn directly to that page to begin reading. Indexed sequential files use exactly the same principle. As shown in Figure 16.3, in indexed sequential file organization there are actually two files for every data file - the data file which contains the records stored in the file, and the smaller index file which contains the key and disk address of each record stored in the data file. The records of the file can be stored in random sequence but the index table is in sorted sequence on the key value. This provides the user with a very powerful tool. Not only can the file be processed randomly, but it can also t processed sequentially. A request for a particular record is first directed to the smaller, more accessible index file, which in turn, points to the physical location of the desired record on the direct access storage device such as disk. For instance, to locate the record of an employee (in the example of Figure 16.3) whose employee is 0004, the computer first searches the index file for this employee code key and obtains the address value 1002 corresponding to it. It then directly accesses the record stored at the address location 1002 of the storage device. This technique of file management is commonly referred to as the Indexed Sequential Access Method (ISAM) and the files of this type are called ISAM files. Employee Address Code (key) Location
0001 0002 0003 0004 •••
1003 1001 1004 1002 •••
Address Location 1001 1002 1003 1003 •••
Employee Record 0002 R. S. Patel... 0004 R. K. Rana... 0001 K. P. Sinha... 0003 N.P.Singh... •«•
Advantages of Indexed Sequential Files Indexed sequential files typically have the following advantages: 1. They can be quite efficiently used for sequential processing for applications in which the activity ratio is high. 2. They can also be used quite efficiently for direct access processing for applications in which the activity ratio is low. Disadvantages of Indexed Sequential Files 1. They require relatively expensive hardware and software resources because they must be stored on a direct-access device such as a disk. 2. They require more storage space than other types of files because the index file can also become quite large. 3. When used for direct access online applications, access to records may be slower than direct files. File Utilities File utilities consist of routines, which perform a variety of generalized operations on data files. Normally, they are quite general in nature in the sense that they can operate on any data format and even on data files stored on different types of storage medium. The operations performed by some of the commonly used file utilities are described below. Sorting
A file sorting utility is used to arrange the records of a file in some defined sequence. This sequence is determined by the ordering of certain specified fields (known as keys) within the record. The simplest case of sorting is an ordering of the records in a file on a single key. For example, the records of an employee file may be sequenced by ascending order of employee code as shown in Figure 16.4. A more complex ordering may be produced by introducing a further key in the sorting process. For example, suppose each record of the employee file also contains a field for department code to which the employee belongs. Now the order of sorting may be employee code within department code. This means that all records for the lowest department code are presented for each employee belonging to that department in ascending sequence of employee code, then all records for the next department code, and so on. This is shown in Figure 16.5. Out of the two keys used in this sorting example, department code is called the primary key and employee code is called the secondary key because the order oi sorting is employee code within department code. The size of keys, the number of keys, and the type of ordering (ascending, descending), which can be specified as input parameters, varies from one sorting utility to another. These parameters decide the extent and sophistication of sorting utilities. A sorting utility enables users to sp6cify their particular sequencing requirements by means of simple parameters. The user specifies the sort keys, and also details about the type of file such as storage device, file labels, record structure. The sorting utility reads the unsequenced input file, and by means of various copying techniques ultimately produces an output file, which contains the records of the input file ordered in the required sequence. Searching A file searching utility is used to find a particular record from a file. The searching is carried out with respect to certain specified fields (known as keys) within the record. For example, in the employee file of Figure 16.4, the user can specify the value 202 for the employee code field to request the computer to search the corresponding employee's record. The efficiency of a search algorithm depends on the file organization. For example, to search a particular recon in a sequential file, the file is scanned sequentially beginning with the first record and the specified key i compared oneby-one with the key field of each record. The search process terminates when the record with the matching key is found. On the other hand, direct or index sequential file organizations enable immediate access to the desired record with the help of either the hashing algorithm (in case of direct file organization) or the index file (in case of index sequential file organization). The user needs to only specify the file and the key, and the searching utility searches through the file and
produces the desired record. Normally, the time required to search a particular record from a direct or index sequential file is much less as compared to the time required to search it from a sequential file. Merging A file merging utility is used to combine the records of two or more ordered (sorted) files into a single ordered file. Each of the constituent files must be in the same order, although the record layout of files need not be identical. The output file produced by the merging utility has the records in the same order as the input files, placing records from each in their correct relative order. Figure 16.6, illustrates the merging of two input files A and B to produce an output file C. Copying A file copying utility is used to produce an exact copy of a file, either from one unit of a storage device on to another similar unit (such as from one type reel to another or from one floppy disk to another), or from one storage medium to another (such as from tape to hard disk or from CD-ROM to hard disk). File copying utilities are often used for taking back-up copies of files. For example, a file may be copied from a hard disk to a tape or floppy for back-up purpose. File copying utilities are also known as peripheral interchange programs since they may be used to copy a file from one peripheral device on to another peripheral device. Printing A file printing utility is used to print a file on a printer to produce hard copy of its contents. Printing utilities often provide the facility to print file contents in different formats. They do often provide some selection and editing facilities to enable parts of files to be printed (such as specified number of records and only certain fields of records). Special printing facilities are often provided to print files that contain program instructions rather than data. Maintenance A file maintenance utility is used to selectively copy the data of one or more files into a new file, or to selectively update a file's contents. For example, facilities provided in a file maintenance utility may include the combining of data from more than one file into a single file, the deletion of records identified by record key or record count within a file, and the selection of specific fields of records to be copied from an existing file into a new file.
DATABASE MANAGEMENT SYSTEM In the database-oriented approach of organizing data, a set of programs is provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their data in a database. All these programs together form a Database Management System (DBMS). The commonly supported features in a typical database management system are described below. Database Models We saw that multiple related files are integrated together to form a database. A database model defines the manner in which the various files of a database are linked together. The four commonly used database models are (they are also known as database structures or database structuring techniques): 1. Hierarchical, 2. Network, 3. Relational, and 4. Object-oriented They are briefly described below. Hierarchical Databases In a hierarchical database, the data elements are linked in the form of an inverted tree structure with the root at the top and the branches formed below. Below the single-root data element are subordinate elements, each of which, in turn, has one or more other elements. There is a parent-child relationship among the data elements of a hierarchical database. A parent data element is one that has one or more subordinate data elements. The data elements that are below a parent data element are its children data elements. There may be many children elements under each parent element, but there can be only one parent element for any child element. Note that the branches in a tree structure are not connected. An example of a hierarchical database structure is shown in Figure 16.7. The hierarchical database structure has been used in a large number of database applications because the data elements of many applications can be neatly organized in the form of a hierarchical tree structure. However, the main limitation of a hierarchical database structure is that it does not support flexible data access because the data elements can be accessed only by following the path down the tree structure. Hence it is important to determine all types of relationships among
the data elements and map them properly in the form of a hierarchical tree structure when the database is first designed. For example, in the hierarchical database of Figure 16.7, an organization maintains a hierarchical database of all its employees. Employees are categorized by the department in which they work, and within a department they are categorized by their job function, such as managers, engineers, technicians, and support staff. If the personnel department has a shortage of support staff one day, producing a list of all support staff, to take a decision for using some support staff from other departments for this department, would not be directly possible. Instead, the support staff assigned to each department would have to be determined department-by-department. Network Databases A network database structure is an extension of the hierarchical database structure. In this model also, the data elements of a database are organized in the form of parent-child relationships, and all types of relationships among the data elements must be determined when the database is first designed. In a network database, however, a child data element can have more than one parent element or no parent at all. Moreover, in this type of database, the database management system permits the extraction of the needed information by beginning from any data element in the database structure instead of starting from the root data element. An example of a network database structure is shown in Figure 16.8. The example shows a network database of a college that maintains the relationships among the courses offered and the students enrolled for each course. As can be seen from the example, the parent and child elements can have many-to-many relationships in a network database structure. That is, each student may be enrolled in several courses, and each course may have a number of students enrolled for it. With this database structure used for this example, it is a course-wise students-wise courses report (a report showing all courses taken by each student). For example, it is easy to tell from the databas structure that in this semester, Maths course has been taken by Seeta, Geeta, Ram and Sohan, and it is also easy t tell that in this semester, Geeta has taken the courses Hindi, Maths and Computer Science. The example also shows a child element that has no parent element (the student named Raju he might be a research student who has not taken any course in this semester). Relational Databases In a relational database, the data elements are organized in the form of multiple tables with rows and columns. Each table of the database is stored as a separate file. Each table column represents a data field, and each row a data record (also
knows as a tuple). The data in one table (or file) is related to data in another table through the use of a common field. An example of a relational database structure is shown in Figure 16.9. The example shows a sample library database comprised of three tables. The first table contains the data of library members, the second table contains the data of borrowed books, and the third table contains the data of books in the library. Each new member assigned a membership number and issued a library card with this number on it. The member's name, address, and membership number are added to the first table. When a member borrows a book, the membership number of the borrower, the book's ISBN number (which serves as the unique identification number for a book), and the due date for returning the book are added to the second table. The first and second tables are related through the 'Membership No.' common field, and the second and third tables are related through the "Book No.' common field. Now suppose the librarian wants a report of overdue books as of 10-11-2001, and wants the list to contain the borrower's details and the book's details for each overdue book. To produce such a report, the database search routine will first search for the due date field in the borrowed books data table to surface the three overdue books. The database management system will then use the 'Membership No.' field to cross-reference the borrower's details for each overdue book from the members data table, and the 'Book No.' field to cross-reference the book's details of each overdue book from the books data table. The final report thus produced would look something like the one shown in Figure 16.10. The relational database model provides greater flexibility of data organization and future enhancements in the database as compared to the hierarchical and network database models. For many applications, the data elements can be organized in the relational database model in a manner that is identical to the actual relationships that exist between the data elements. Hence many users find the relational database model to be closer to their intuitive model of the real-life situation, making this model easier to understand and use. Moreover, if new data is to be added to an existing relational database, it is not necessary to redesign the database afresh. Rather, new tables containing the new data can be added to the database, and then these tables can be related to the existing tables through the use of common key fields. Data can also be reorganized when necessary to create new tables by selecting certain rows or specific columns from other tables, or by joining columns and rows from two separate tables. Object-oriented Databases
The hierarchical, network, and relational database models (henceforth referred to as conventional database models) were developed for the conventional business data processing applications, such as inventory control, payroll, accounts, employee information systems, and so on. Attempts to make use of these database models in a wide variety of other types of applications exposed several serious shortcomings of these database models. These applications include computeraided design (CAD), computer-aided engineering (CAE), computer-aided manufacturing (CAM), computer-aided software engineering (CASE), expert systems, and multimedia systems, which deal with documents having text, graphics, animation, audio, and video. Some of the key features required for effective modeling of these applications that are absent in conventional database models are: 1. Ability to model complex nested entities, such as design and engineering objects, and multimedia documents. Conventional database models do not provide mechanisms, such as configuration management, to represent and manage such entities. 2. Support for general data types found in object-oriented programming languages. Database management systems based on conventional database models support only a limited set of atomic date types, such as integer, string, etc. They do not even allow the storage and retrieval of long unstructured data such as images, audio, and textual documents. 3. Support for frequently useful object-oriented concepts such as object, class, inheritance, etc. (these concepts have been explained in Chapter 12). The objectoriented concepts consist of a number of data modeling concepts, such as aggregation, generalization, and membership relationships, which are very useful in modeling complex applications. Database management systems based on conventional database models do not provide the necessary functions for implementing object-oriented concepts. This means that applications programmers must explicitly represent and manage objects and the relationships among the objects in their programs. 4. Support for proper match between object-oriented programming languages and database languages A database application is normally implemented by using some conventional programming language (such as COBOL, PASCAL, PL/1, C, or C++) and some database languages (data definition language, data manipulation language, query language), which are part of the database management system. With the popularity of object-oriented paradigm, the use of object-oriented programming languages for implementing applications has become a common practice. However, database languages used in database management systems for conventional database models do not use object-oriented concepts for implementing applications. This mismatch between object-oriented programming languages and database languages used in database management systems for conventional database models, makes database implementation of many applications inconvenient.
The object-oriented database model was introduced to overcome the above listed shortcomings of conventional database models. A data model is a logical organization of the real-world objects (entities), constraints on them, and relationships among objects. A database model that captures object-oriented concepts such as class hierarchy, inheritance, and methods is ah object-oriented database model. An object-oriented database is a collection of objects whose behavior and state, and the relationships are defined in accordance with an objectoriented database model. An object-oriented database management system is one, which allows the definition and manipulation of an object-oriented database. The object-oriented database model is the preferred model for a large range of database applications such as CAD, CAE, CAM, CASE, expert systems, and multimedia systems. The reason is that object-oriented concepts already include such data modeling concepts as grouping of objects into a class, aggregation relationships between an object and other objects it consists of (nested objects), and generalization relationships between a class of objects and classes of objects specialized from it. The fact that an object-oriented database model includes the aggregation and generalization relationships means that an object-oriented database management system provides direct support for the definition and manipulation of the relationships among objects. This in turn means that application programmers need not explicitly program and manage these relationships. An example of an object-oriented database structure is shown in Figure 16.11. The class Vehicle is the root of a class-composition hierarchy, which includes the classes VehicleSpecs, Company, and Employee. The class Vehicle is also the root of a class hierarchy involving the classes TwoWheeler and FourWheeler. The class Company is in turn the root of a class hierarchy with subclasses DomesticCompany and ForeignCompany. It is also the root of a class-composition hierarchy involving the class Employee. For the database structure of Figure 16.11, a typical query may be like "list out the President's and Company's names for all the companies located in Pune, India that manufacture two-wheeler vehicles." Main Components of a DBMS A DBMS frees the users of a database from the need to worry about the organization and location of data. It allows users to organize, process, and retrieve selected data from a database without knowing about the underlying database structure. The four major components of a DBMS that enable this ease of use of a database are:
1. Data Definition Language (DDL), 2. Data Manipulation Language (DML) 3. Query Language, and 4. Report Generator. They are briefly described below.
-
Data Definition Language (DDL) The data definition language (DDL) is used to define the structure of a database. The database structure definition (also commonly known as database schema) typically includes the following: 1. Defining all data elements to be included in the database. 2. Defining the organization of the data elements (fields) into records (or tuples), tables, etc. 3. Defining a name, field length, and field type for each data element. A field name is used to refer to the corresponding data element while accessing it. Field length is used to define the maximum size c the data element (such as an employee name field, may have a maximum size of 30 characters Common field types are numeric (can contain only the digits 0 to 9, a decimal point, and a + or sign), alphanumeric (can contain a combination of alphabetic characters, special symbols, and digits logical (can contain one of two possible values - Yes/No or True/False), date (can contain a single date), and memo (can be used to hold any type of reminder that the user might like to type). 4. Defining controls for fields that can have only selective values. For example, in an employee database, the sex field can have controls to accept only the values M or F. Similarly, the age field can have controls to accept only the values 18 or more and less than 70. Such controls ensure the correctness of entered data to some extent. 5. Defining access controls to the various tables, records, and fields for different categories of users to protect the privacy of sensitive data items from unauthorized users. 6. Defining the logical relationships among the various data elements of the database. In short, virtually everything about the database structure is included in its schema. It is possible to generate complete description of a database from its schema. This description, in turn, can be used by systems analysts ii defining new applications on the database. Database systems are typically installed and coordinated by an individual called the database administrator. He/she has the overall authority to establish and control data definitions and standards. He/she is responsible for determining the relationships among data elements, and for designing the database security system to guard against unauthorized use. He/she also trains and assists applications
programmers in the use of the database. A data dictionary is developed and used in a database to document and maintain the data definitions. It is automatically created/updated by the DDL module of the DBMS as the database schema is defined or changed. Data Manipulation Language (DML) Once the structure of a database (database schema) has been defined, the database is ready for the entry and manipulation of data. The data manipulation language {DML) includes all the commands that enable the users to enter and minipulate the data. With these commands, the user can do such things as add new records to the database, navigate through the existing records, view the contents of the various fields of a record, modify the contents of one or more fields of a record, delete an existing record, and sort the records in a desired sequence. In some DBMS, the data definition language and the data manipulation language are combined together, while in others, they are supported as separate DBMS components. Query Language Although it is possible to navigate through a database one record at a time to extract the desired information from the data stored in the database, this approach can be very inefficient and frustrating when there are thousands of records and several files in the database. Hence all database management systems provide a query language that enables users to define their requirements for extracting the desired information from the database in the form of queries. For example, from an inventory database, a user may be interested in such information as "list out the item description and vendor name for all items whose current inventory level is less than 20 units" or "list out the stock number and item description for all items with a profit margin greater than 25%". Earlier each database management system had its own query language. The main problem with this was that queries developed for one DBMS could not be used with other DBMSs. However, eventually one query language, called SQL (pronounced "S-Q-L"), emerged as an industry standard. It was originally developed by IBM and was based on an earlier query language called SEQUEL. SEQUEL being an acronym for "Structured English QUEry Language", many people pronounce SQL as "Sequel" and believe that it stands for "Structured Query Langauge". However, SQL does not stand for anything. Today, SQL is the standard query language used in many DBMSs.
A query language can be easily learnt by a non-programmer. This enables normal users of a database to access the database for desired information without the help of any programmer. Report Generator A report is the presentation of information extracted from a database. The report generator enables the users of a database to design the layout of a report so that it can be presented in the desired format. This means that the user can specify proper spacing between the data items to be presented in a report, and can also include suitable report titles and subtitles, column heading, page numbers, separation lines, and other elements that make a report more readable and presentable. The report generator can also be instructed to perform arithmetic operations (such as calculating subtotals and totals) on the data found in numeric fields to make a report more meaningful and useful. A user can define the layout of a report and then store it for later use. This facility is often used for generating periodic reports. For example, this facility is found to be very useful for generation of employee pay-slips from an employee database every month, or generation of consumer bills every month, for the consumers of a service. The report layout format remains the same every month, only the data changes. Creating and Using a Database In this section we will have a closer look at how a database is created and used. Large databases on large computer systems are normally created and maintained by professional programmers. The users of these databases, however, need not be programmers. They can be easily used by non-programmers to access the data and produce reports. On the other hand, many database management systems designed for personal computer systems enable non-programmers to not only use but even to create their own databases. These databases are normally simple in structure and small in size. Creating a Database Creation of a database is a three steps process: 1. Defining its structure, (schema), 2. Designing forms (custom screens) for displaying and entering data, and 3. Entering the data into it. These steps are briefly described below.
Schema Definition The first thing we do to set up a database is to define its structure (schema definition). This is done by identifying the characteristics of each field in it. A good way to begin defining the schema of a database is to list down on paper all the fields that are to be included in the database, and then to determine the name, type, and size of each field. This information is then captured into the system using a tool called schema builder. The schema builder enables a user to define a database schema interactively by prompting the user to enter the field name, field type field size, and so on for each field. While defining the schema of a database, it is important to consider possible future needs and the needs of all types of users of the database. That is, all possible fields that may be needed should be included in the database structure while defining it. Although it is possible to modify the database structure at any time, making such modifications is a time-consuming process. Hence it is always better to design a database carefully in the first instance and minimize the need to modify the database structure. Designing Forms After defining the structure of a database, the next step in database creation is to design the forms (custom screens) that permit convenient data entry. Each form displays a set of fields of the database structure with appropriate amount of blank spaces against each to enable data entry for those fields. To facilitate easier data entry, forms are often designed with several fancy features such as: 1. A list box for a field that lists several options on it from which the user can select a choice. For example, in the form of Figure 16.12, the SEX field may have a list box that lists the options "Male" and "Female" on it and the user simply selects the appropriate option for an employee. Depending on the user's selected option, the system automatically enters "M" or "F" in the SEX filed. Similarly, the STATE field may have a list box that lists all the states on it from which the user can select the appropriate option for an employee's address. 2. Simple validation checks defined for a field to ensure correctness of entered data to some extent. For example, in the form of Figure 16.12, a validation check may be defined for the POSTAL CODE field to accept only six numeric characters. With this, the field will not accept less than or more than six characters. Similarly, a validation check may be defined for the AGE field to accept values only in the range of 18 to 70 (both inclusive). Validation checks can also be used to force a certain character position of a field to be a letter or numeral.
3. Automatic conversion of typed characters to upper-or lower-case. For 16.12, this feature may be used with the STATE field. Hence, the system will accept “mh”, “Mh”, "mH" or "MH" for the state code of Maharashtra, and will automatically convert the entry to "MH”. This feature can greatly ease data entry and ensure the uniformity of data. 4. Automatic formatting of certain fields. For example, in the form of Figure 16.12, this feature can be used with the TELEPHONE NO. field to automatically display the value of this field in the specified format (with the parentheses, space, and hyphen). That is, to enter the telephone number "(020) 5680-489", the user only needs to type "0205680489", and the system automatically causes the form to display "(020) 5680-489". Entering Data After the forms have been designed, the database is ready for entry of data. Data is entered one record at a time. To enter the data, the user issues a command that calls up and displays the appropriate form with blank fields. The user then keys in the data for each field in the appropriate spaces. In this manner, the user enters the data for the first record, then for the second record, and so on. In most database systems, the records are automatically assigned a number as they are entered. While entering data into the fields, the tab key or enter key is usually used to move to the next field. Pressing enter or tab key in the last field on the form saves the record in the database and moves to a new, blank form the next record to be entered. In addition to using the tab or enter key to move forward through fields, one can directly go to any field on the form at any time by clicking on it with the mouse. Viewing, Modifying, Deleting and Adding Records All database systems .provide commands to view, modify, delete, or add the records of an already established database. The command for viewing a record enables the user to display the data in the various fields of a record in the same screen format as that used for data entry. The user can specify the record to be displayed by specifying its key field value. The database system usually also provides the flexibility to the user to move between records for viewing different records, such as "go to the first record", "go to the previous record", "go the next record", and "go to the last record". Many database systems also provide a facility to set up a filter that allows the user to browse through and view only those records, which meet some criterion. For example, in the employee database created by using the form of Figure 16.12, if the user wants to view the records of only female employees, the user can set a
filter for the "Sex" field and only the records that contain "F" in that field will be displayed. Note that while a filter is set, to user cannot access the records that do not meet the filter criteria. Filters provide a quick and convenient way narrow down the number of records the user has to work with. The command for modifying a record enables the user to not only view but also to update the data in the various fields of a record. To modify the contents of a particular field of a record, the record is first displayed, then the cursor is positioned in the field at the appropriate character position where the change is to be made by clicking the mouse there, and then the contents of the field is appropriately edited. The data in any other field of the record can be similarly edited. Finally, the enter key has to be pressed for the changes to take effect. Some database systems may prompt the user to confirm the changes before effecting the changes and allowing the user to mo to another record. The command for deleting a record enables the user to remove the selected record from the database. To delete record, the user first selects the record either by specifying its key field value or by using the facility to mo between records just as is done in the case of viewing a record. The user then uses the delete command, and the record is deleted. Most database systems prompt the user to confirm the deletion of the record before deleting it. This feature prevents deletion of a record by mistake. The command for adding a record enables the user to add new records to the database. When this command is enabled, the system displays a blank form and waits for the user to enter data. The user then keys in the data for each field in the appropriate spaces and finally presses the enter or tab key after keying the data in the last field to save the record. On doing this, the system displays a new blank form for the next record to be entered. If no more records are to be added, the user can use the mouse to select the option to terminate the data entry process for adding new records. Searching for Desired Information A database management system derives much of its power from the ease with which its users can search for the desired information from the large volume of data stored in a database. That is, users can search a database with great ease for obtaining answers to a wide range of queries. The commonly supported features in modern database systems for this are: 1. Find command, 2. Query language, and 3. Query By Example (QBE) These features are briefly described below.
Find Command The find command is used for simple database queries, like searching for records having a particular string pattern in a field. For example, in the employee database created by using the form of Figure 16.12, the Find command may be used to list out the records of all employees whose last name is "SINHA". Similarly it may be used to list out the records of all employees who belong to the city of "PUNE". To use Find, the user has to type the string pattern to be searched and then has to indicate which field to search in. For instance, in the above example, the user has to type "STNHA" and indicate that this string has to be searched in the "LAST NAME" field. The user can specify either a single field or all fields. The Find command cannot be used for creating an elaborate set "of criteria for complex queries. Furthermore, it can operate on only one table at a time, and the user cannot save a specified criteria for future use. Query Language For handling complex quieries, all database systems support a query language. Most of these query languages conform to the SQL standard. In SQL, the user has to specify the criteria for search along with the fields and table (or tables) with which to work with. The criteria for search can be built by using the relational operators (= [equal to], > [greater than], < [less that], and combinations of these operators) and the logical operators (AND, OR, and NOT). For example, to list out the names of all employees whose last name starts with the letter "S", who belong to "PUNE", and whose age is more than 40 years, the SQL query will look as follows: SELECT [LAST NAME], [FIRST NAME], [MIDDLE NAME] FROM Employee WHERE ([LAST NAME] = "S..") AND (CITY = "PUNE") AND [AGE > 40]) The keywords SELECT, FROM, and WHERE tell the SQL engine how to interpret each part of the query statement. The SELECT keyword tells SQL which fields are to be displayed for the records that match the criteria. The FROM keyword tells SQL which table(s) to work with. The WHERE keyword tells SQL the criteri for selecting the records (search criteria). The brackets [...] around some of the field names are needed in the above example because these field names contain spaces, and the brackets help the database to interpret each field name correctly.
A query language can be easily learnt and used even by a non-programmer because the complexity of a query language statement is normally more or less of the same order as given in the above SQL statement. Furthermore, a query language uses only a few keywords which are easy to remember and use. In fact, SQL has only a few dozen or so basic keywords. Other advantages of using a query language are that a query statement can be used for creating an elaborate set of criteria for complex queries, it can operate on multiple tables at a time, and a specified criteria can be saved for future use. Query By Example (QBE) Although query languages are easy to learn and use, many database developers further simplify the job of database users of specifying search criteria by providing front ends for the query language that collect the facts about a query from the user and compose query language statements internally. The front end usually consists of a form (called QBE form) and the user simply specifies the search criteria by inputting values into the fields of this form. Again the values may be input either by typing them or by selecting an option from a set of options provided for a particular field, depending on how the front end has been designed to work. The QBE form is designed to collect all necessary information from the user for composing query language statement(s) for the search criteria. Once the user completes the QBE form, the QBE engine automatically converts the user inputs into suitable query language statement(s) for search processing. Hence the user is relieved of remembering the query language keywords and using them with the correct syntax to form queries. This front-end feature is called query by example (QBE) and has been found to be very userful for many database users, especially the beginners. Creating Reports The user of a database system can use the report generator of the database system to assemble the output of a database query in the desired format. For this, the user creates a report specifying the layout of the display (or printout) of the fields requested by the user in the query. The user can also specify the titles and subtitles for the report, the column headings for the various fields, and other elements to make the output appear more presentable. Furthermore, the user can even specify the output to be sorted with respect to one or more fields in the output. When the sorting is on more than one field, the user has to specify the primary, secondary, and tertiary key fields. A created report can be saved and used later for generating similar reports whenever required.
Figure 16.13 shows a sample output of a report created for the employee database whose data entry form is shown in Figure 16.12. The report is used to display (or print) the list of employees who belong to the city of PUNE. The report contains the last name, first name, address-1-, address-2, and telephone number of all such employees. The report is sorted to present the list of employees in alphabetical order of their last name. Many database developers further simplify the job of database users by creating commonly anticipated reports, storing them, and providing a simple graphical user interface to select from the available reports. In this case, the user has to only select one of the reports from the available options while outputting query results. The user needs to create a new report only when none of the available options meet his/her requirement. The newly created report can then be added to the list of available Options for future use. LIST OF EMPLOYEES WHO BELONG TO PUNE DATE: DECEMBER 15, 2001 LAST FIRST ADDRESS-1 ADDRESS-2 NAME NAME Gupta Rajiv A-12, Nandanvan M. G. Road
TELEPHONE NUMBER 4623-489
Murli Pandey Raina Ray Sen Singh
5863-490 5865-323 5755-832 4685-635 5762-333 5728-628
Tapan Rupa Pushpa Suhas Prakash Deepak
A-ll, Vrindavan D-18, Vrindana C-15, SaritaVihar B-05, Royal Villa B-16, AnandPark A-22, Anand Park
Pashan Road Pashan Road Aundh Road M. G. Road Aundh Road Aundh Road
Figure 16.13. A sample output of a report to generate a list of employees who belong to Pune. The report is sorted to present the list in alphabetical order of their last name. MULTIMEDIA DATABASE SYSTEMS Traditional database systems are suitable for the storage and management of simple data types such as strings and numbers. With the increased use of computers for multimedia applications over the last several years (such applications are covered in Chapter 19), an increasing trend is to use database technology for the storage and management of multimedia data such as text, graphics, animation, audio and video. The nature of these data types and the hardware and software features required in a system to handle them are given in Chapter 19. The very nature of the multimedia data and the frequent requirement of handling several of them together (such as audio and video data have to be frequently handled together) make traditional database systems unsuitable for the storage and management of multimedia data. This limitation of traditional database systems motivated researchers to build technologies for multimedia
database systems. We saw before in this chapter that a database consists of a collection of related data, while a database management system is a set of programs to define, create, store, access, manage and query the database. Similarly, we can view a multimedia database as a collection of multimedia data items, such as text, graphics, animation, audio, and video, and a multimedia database management system as a set of programs to define, create, store, access, manage, and query the multimedia database. Requirements for Multimedia Database Systems In order to deal with multimedia data, multimedia database systems must meet certain special requirements. These requirements are both from the perspective of special hardware and software requirements for the computer systems to be used for this purpose, and special database features needed for handling multimedia data. As the hardware and software requirements for handling multimedia data have been covered in Chapter 19, here we will discuss the database features needed for handling multimedia data. These features are discussed below. Support for Multimedia Data Storage and Retrieval By their very nature, multimedia data are characterized by huge volumes. A multimedia database should be capable of storing and managing several gigabytes of data for small multimedia data objects such as images, and several hundred terabytes or more if the database must hold significant amounts of video or animation data. To meet the requirement of huge data volumes, multimedia database systems often use a hierarchy of storage devices The storage hierarchy typically consists of primary storage (such as random access memories), secondary storage (such as magnetic disks), and tertiary storage (such as CD jukeboxes, or magnetic tape libraries). In a hierarchical storage system like this, performance of data retrieval depends on the efficiency of the migration mechanisms used to assign the multimedia data items to the optimal level in the storage hierarchy. The requirement of huge storage capacity is often taken care of, to a large extent, by using data compression and decompression techniques (described in Chapter 19). Hence a multimedia database management system typically requires mechanisms for handling huge data storage (possibly comprising of multi-level hierarchy), migrating data automatically from one level to another in the storage hierarchy for efficient data retrieval, compressing and decompressing data at the time of storage and retrieval respectively, and transferring and distributing huge volumes of data across the network. Support for Multimedia Data Types
Multimedia database management systems must provide type support for common multimedia data types. That is, it is not sufficient for a multimedia database system to serve only as a repository of multimedia data, but the database system must also be intelligent enough to understand the data it manages. Hence it must have the capability to recognize and operate on several standard multimedia data formats, such as JPEG, GIF, and MPEG. Internally, a multimedia database management system may support a single (perhaps proprietary) data format for each data type. For example, all types of multimedia data may be treated internally as Blobs (binary large objects), as in simple data repositories that do not recognize or support multimedia data formats. In this case, the multimedia database system must have conversion routines to convert data from original format to internal format on import to the database, and from internal format to original format on export from the database. Some multimedia data types such as video, audio and animation sequences also have temporal (time-related) requirements, which have implications on their storage, manipulation, and presentation. Similarly, images, graphics, and video data have spatial (space-related) constraints in terms of their content. A multimedia database system must be suitably designed to understand and meet these additional requirements of these multimedia data types in order to handle them successfully. Support for Multimedia Data Relationships A multimedia database system supports multiple data types. It is often required to integrate two or more data types to form new composite multimedia data types and present them to the user. Often composition and presentation of multimedia data types requires spatial (space-related), temporal (time-related), and/or synchronization relationships to be established among various data items of different types. Spatial relationships organize the data's visual layout. For example, superimposition of a text object on a specific part of a displayed image/video object. Temporal relationships dictate the temporal layout, ensuring the timely presentation of the right data. A simple example might be, "Play video and audio al, and, when finished, play music ml until slide show s1 is done". Synchronization relationships control the synchronized presentation of two or more objects, such as maintaining lip synchronization between a movie and a sound track. In order to take care of all such multimedia data composition and presentation requirements, a multimedia database system should provide facilities to define and support several kinds of relationships between multimedia data objects. Support for Multimedia Data Modeling
A multimedia database model must haw features to capture the static and dynamic properties of the database contents (multimedia data). The static properties could include the data items that make up the multimedia data, the relationships between the data items, the data item attributes, and so on. Examples of the dynamic properties include interaction between data items, operations on data items, user interaction, and so forth. There is a general consensus that the object-oriented database model is the most suitable model for multimedia databases. In most cases, multimedia data types are implemented as classes, having attributes and methods to support each data type. They may be part of a large class hierarchy or framework. A framework goes beyond a simple type definition to give more complete support for the data types, including several methods and additional supporting classes (to import, view, and export). While type frameworks focus on type-specific support, a database model for multimedia databases must also have features to express several kinds of relationships between multimedia data types for composition and presentation. Hence, a database model for multimedia databases goes much beyond an ordinary object-oriented database model used for modeling traditional databases. Support for Multimedia Data Manipulation A database management system needs to provide support for easy manipulation of the data stored in a database. Implementing features for manipulation of simple data types is simple, but it is not a simple task for multimedia data. Multimedia databases require multimedia data manipulation languages (DMLs) or constructs with features to support not only simple cut or crop operations on multimedia data, but also to ensure that these operations adhere to the spatial, temporal, and synchronization constraints discussed before. For example, to main the temporal constraints, these operations must provide predictable and worst-case information about the operation overhead. This lets the system determine if quality of service (QoS) guarantees can be met when the operation is inserted within the data stream between the source (database server) and the sink (the viewer). Another concept related to data manipulation used in multimedia database management systems is that of derived data. A modified data needs to be stored in the database for future reference. It may either replace the old version or exist as another version of the data. If users constantly create new versions of data, the database size will grow rapidly. To reduce the need to store new versions of data, a multimedia database management system stores the operations to generate the new data (called derived data) rather than storing the new data itself. Support for Multimedia Data Query
Some desired information can be extracted from the data stored in a database by initiating queries. Queries contain predicates that must be satisfied by any data retrieved. It is easy to formulate predicates and look for its matching data in conventional databases. For example, "Find all employees for whom CITY = "PUNE" and AGE > 40". However, this is not an easy task while dealing with multimedia data. The three commonly used methods in existing multimedia database systems for querying multimedia data are: 1. Manual keyword indexing. This method is currently the most popular and straightforward method to query multimedia databases. In this method, descriptive keywords are manually associated with every multimedia data stored in the database. For example, keywords like “long face, sharp pointed nose, broad moustache, no beard, fair complexion, mole on right side of chin” may be used to describe an image data of a face. These keywords are metadata, since they are data about multimedia data. For querying, a user specifies some keywords and the database management system looks for images with matching keywords (the images themselves are not queried). All images with matching keywords are returned as the result of the query. This method has the following limitations: - Keyword classification is subjective since it is performed by a human. - Exceptions will always exist, and some data may be incorrectly classified. - Keywording is usually limited to a well-defined abstraction of the data (for example, for every image of a face, a specific set of features is classified). This means that if the abstraction becomes altered, then all of the data must be reviewed again, adding new keywords as required. This could be a formidable task even for small databases. 2. Content-based retrieval/querying (CBR/CBQ). In this method, queries are based on the content oi multimedia data. When the data enters the database, it is subjected to data analysis to derive certain attributes of the data (called derived data). The derived data could include attributes such as lines, shapes, colors, and textures from which objects could be determined at some later time. Queries against this data then take place on the derived data. For example, "Find all graphic objects having red triangles and green circles." Similarly, "Find all video clips having a flying aeroplane." To distinguish CBR on temporal data such as video and audio, from that on images and graphics. CBR on images and graphics are called static CBR (or SCBR), and CBR on video, animation, and audio are called dynamic CBR (or DCBR).
With both types of CBR, specifying precise queries (exact matching) generally is not practical. Therefore, multimedia database query languages must have features to support fuzzy predicates such as "Looks like this" to find approximate matches. Furthermore, since queries might not yield exact matches, we need facilities for ranking the retrieved results according to how closely they match the given query. With the ranking facility, the multimedia database system should also support the facility to browse the various retrieved data items, and the facility to retrieve similar items based on one or more of the already retrieved items. 3. Query by example (QBE). QBE in case of multimedia database systems is a special user interface to support CBR. These interfaces allow the user to specify the attributes of the desired object(s) by providing examples to match such as, a drawing, photograph, action, or sound. For example, the user can" draw an example of the type of image to be retrieved, by using shape, color, or texture. The query manager then uses that example to find other similar images in the database. Support for Multimedia User Interface The diverse nature of multimedia data requires an equally diverse user interface for interacting with the database. Since multimedia database systems require spatial and temporal rendering of multimedia data, the user interface design must include features to control and devices to render the presentation. This may include simple VCR-type control panels enabling the user to play, fast-forward, pause, or rewind a presentation, as well as more advanced controls including interactive data filtering, querying, and visual perspective controls. In addition, the user interface should also have features to allow the user to control the quality of service (QoS) parameters to adjust for resource limitations, cost of delivery, and personal visual and aural preferences. KNOWLEDGEDISCOVERYINDATABASES_(KDD) The tremendous success of database technology has resulted in a huge amount of data stored in databases in all types of industries and organizations. The amount of data being collected and stored in databases continues to grow at a phenomenal rate. A question that has become common for the owners of many databases is "What do we do with so much of gathered data? Should we discard several years old data and keep only the recent ones?" But people have realized that old data should not be discarded because it contains some valuable information implicitly coded within it. Organizations have begun to realize that the volumes of data they have been storing contain an important, and largely ignored, resource - broad knowledge about their businesses that could support better business decisionmaking if properly exploited. For example, data collected in scientific databases if properly analyzed can provide useful observations carefully collected about some
phenomenon under study; data collected in business databases if properly analyzed can provide useful information about critical markets, competitors, and customers; and data collected in manufacturing databases if properly analyzed can provide useful information about performance and optimization opportunities, as well as the keys to improving processes and troubleshooting problems. The traditional method of turning data into knowledge relied on manual analysis and interpretation. However, manual analysis of data is becoming impractical in many domains as data volumes grow exponentially. Furthermore, manual analysis of a dataset is slow, expensive, and highly subjective. When the scale of data analysis grew beyond human capabilities, people started looking to computer technology to automate the process. This initially resulted in the use of statistical techniques on computers for automatic data analysis. Although the use of statistical techniques along with file management tools once sufficed for analysis of data in databases, the large size of modern databases, the mission-critical nature of the data, and speed with which analyses need to be made, motivated researchers to look into new approaches of data analysis. This ultimately resulted in a new generation of techniques and tools for automatic analysis of huge volumes of data. These new techniques and tools meant for extraction of knowledge from large datasets are the subject of the field called knowledge discovery in databases (KDD). The KDD Process KDD is an umbrella term describing a variety of activities for making sense of data. It is used to describe the overall process of finding useful patterns in data, including not only the data mining step of running specific knowledge discovery algorithms but also pre- and post-processing of data and a host of other important activities. The KDD process, shown in Figure 16.14, typically consists of the following steps: 1. Data Selection. This step involves selection of a dataset or focusing on a subset of variables or data samples on which knowledge discovery is to be performed. This step is more significant than it may appear to be, especially when the data is to be pulled from multiple data sources. For data selection, data often needs to be extracted from different databases and joined, and perhaps sampled. 2. Data Preprocessing. After the selection of data, the selected data needs to be preprocessed (cleaned). Data preprocessing/cleaning involves looking for obvious flaws in the data and removing them, and removing records with errors or insignificant values. Removal of flaws in the data involves collecting the necessary information to model or account for the flaws, and deciding on strategies for handling missing and unknown values in data fields. Data cleaning also involves data conversions to ensure uniform representation of data in all records. This is mainly required when the target data is derived from different
databases in which values for the same field may be represented using different notations. For example, the sex field may be represented as 1 for male and 0 for female in one database, and M for male and F for female in another database. The process of data cleaning is often referred to as data scrubbing. 3. Data Transformation. After the selection and cleaning process, certain transformations on the preprocessed data may be necessary. These range from conversions from one type of data to another, to deriving new variables using mathematical or logical formulae. This step also involves finding useful features to represent the data, depending on the goal of the knowledge discovery task, and using dimensionality reduction or transformation methods to properly summarize the data for the intended knowledge discovery. 4. Data Mining. This step involves application of the core discovery procedures on the transformed data to reveal patterns and new knowledge. This includes searching for patterns of interest in a particular representational form or a set of such representations, including classification rules or trees, regression, clustering, sequence modeling, dependency and line analysis. All this requires selection of method(s) to be used for searching for useful patterns in the data, such as deciding which models and parameters may be appropriate and matching a particular data mining method with the overall criteria of the KDD process. 5. Data Interpretation/Evaluation. This step involves interpreting the discovered patterns, as well as possible visualization of the extracted patterns. In this step, redundant or irrelevant patterns are removed, and powerful visualization tools are used to translate the useful patterns into the form most understandable by the users. The user then incorporates this knowledge into the performance system, taking actions based on the knowledge, or simply documenting it and reporting it to interested parties. It may be noted from Figure 16.14 that a realistic KDD process is not simple and linear, but thoroughly iterative and interactive. That is, the results of analysis are fed back into the modeling and hypothesis derivation process to produce improved results on subsequent iterations. It may also be noted here that often people use the terms data mining and KDD interchangeably because data mining is key to the KDD process. However, as shown in the figure, the data mining activity (which is often considered as the core activity of the KDD process) takes only a small part (estimated at 15% to 25%) of the effort of the overall KDD process. The activities from data selection to data transformation in the entire KDD process are popularly known as data warehousing. As data warehousing and data mining have gained wide popularity, they are separately covered below. DATA WAREHOUSING
Data warehousing is the process of creating a repository of integrated data for the purpose of decision support and analysis. A data warehouse is the resulting repository of integrated data. It is a subject-oriented, integrated, time-variant, and non-volatile collection of data in support of management's decision-making process. The meaning of the key terms used in this definition is as follows. 1. Subject-oriented. This means that all data in a data warehouse is organized around a specific subject area from the perspective of the end user (such as customers, products, or accounts) to rapidly provide information relating to this specific area. This is in contrast to most operational database systems in which data is organized around the specific functions or activities of an application (such as, inventory control, pay-roll, or invoicing) to enable efficient access, retrieval, and update. 2. Integrated. Data residing in many different operational database systems is often inconsistent in terms of coding and formats. On the other hand, even when the data in a data warehouse is sourced from multiple database systems (or multiple data sources), it is ensured during the data clearing process that all data in the data warehouse is consistent and uniform, thereby eliminating the issues of data incompatibility and portability. 3. Time-variant. In operational database systems, data is accurate at the moment of access only. However, all data in a data warehouse is accurate at a specific point-in-time and can be kept over a long time period. Due to this feature, the data in a data warehouse provides a historical perspective. 4. Non-volatile. In operational database systems, data is continuously updated on a record-by-record basis. On the other hand, in a data warehouse, data is periodically loaded en mass and accessed later. It is never updated and retains its integrity. Thus, data in a data warehouse is static, not dynamic. The only operations that occur in data warehouse applications are the initial loading of data, access of data, and refresh of data. Due to this reason, the physical design of a data warehouse optimizes the access of data, rather than focusing on the requirements of data update and delete processing. Need for a Data Warehouse A data warehouse is typically needed in an organization in one or more of the following situations: 1. The organization has several operational database systems, but there is no mechanism to present a holistic view of the entire gamut of information from the data stored in these databases. 2. Same data is present on different operational database systems in different departments about which users are not aware and/or different systems are giving different answers to the same business questions.
3. Decision makers and policy planners are not able to obtain easily decipherable, timely, and comprehensible information due to the non-availability of sophisticated tools. 4. IT or business staff are spending a lot of time developing special reports for decision makers. Building a Data Warehouse As shown in Figure 16.14, a data warehouse is typically built by using the following processes: 1. Data selection 2. Data preprocessing/cleaning 3. Data transformation A description of these processes has already been presented before. The transformed data is structured in the form of multi-dimensional cubes in the data warehouse so that different types of users (say from different department of an organization) can view the same data in different form depending on their interest. The data transformation tools are known as OLAP (On-Line Analytical Processing) tools. Finally, data access tools allow the end users (decision makers) to directly and easily access the desired information from the data warehouse without the help of any IT staff. The overall architecture of a typical data warehouse is shown in Figure 16.15. Few Related Terminologies Some commonly encountered terminologies while dealing with data warehousing are briefly described below. OLTP and OLAP OLTP stands for On-Line Transaction Processing. It is the database architecture designed specifically for transaction processing. OLTP describes systems that support day-to-day business operations. These are transaction processing systems, with many users concurrently accessing a database to insert, retrieve, update, and delete individual records. Hence OLTP systems are optimized for the entry, retrieval, and update of individual pieces of fact, not for the support of complex, large-scale data analysis. On the other hand, OLAP stands for On-Line Analytical Processing. It is the database architecture designed specifically for effective data analysis. OLAP describes systems that integrate data from one or more systems, then aggregates
and transforms the data to support effective decision-making. OLAP systems are based on the concept of multi-dimensional databases and allow the users to analyze the data using elaborate, multi-dimensional complex views. In short, OLAP helps an organization to decide what to do and OLTP helps the organization to do it. Figure 16.16 lists out the differences between OLTP and OLAP systems. Features OLTP OLAP Stands for Designed for
On-Line Transaction Processing On-Line Analytical Processing Supporting day-to-day business Supporting effective data analysis operations Used for Implementing the policy decisions Making policy decisions regarding what to do Structure Transaction processing (entry, Complex, large scale data analysis optimized for retrieval, and update of individual pieces of fact) Data model Normalized Multi-dimensional Nature of access Insert, retrieve, update, and delete Retrieve summarized information and records load data in batches Access tools Query language Query language and advanced analytical tools Type of data Data that runs the business Data to analyze the business stored Nature of data Detailed Summarized and detailed Performance Transaction throughput Query analysis throughput metric Figure 16.16. Differences between OLTP and OLAP systems. Data Mart A data mart is a subset of an organization's data warehouse. It contains data that is of value to a specific business unit or department rather than the whole organization. It can contain both detailed and summarized data for each specific business areas of an organization. The data may be captured either from the organization's operational systems or from the organization's data warehouse. A data mart can be used to analyze data only to an extent of single business area or department unlike the organization's data warehouse, which can be used to analyze da across multiple business areas or departments of the organization. A data mart can be built at a much lower cost, time, and effort than that involved in building a data warehouse. Because of their low cost and shorter development time, data marts start giving an early payback, thus justifying the return of investment.
Metadata An important component of every data warehouse is the metadata. Simply stated, metadata is data about data. It is a catalog of information that keeps track of what is where, in the data warehouse. The metadata of a data warehouse typically stores the definitions of source data, target data, source to target mappings, transformation information, and update information. It is an integral part of a data warehouse, which must be constructed along with the data warehouse. Metadata is to a data warehouse what road map is to a navigator. It is the key to providing users and developers with a road map to the information in the data warehouse. Drill Down and Roll Up Analyses Information from a data warehouse can be extracted and viewed as per the requirement of the end user. Both aggregated as well as detailed information can be arrived at, depending on the analysis being carried out on the data. The process of viewing information from a higher level (in the hierarchy) to a more detailed view of information is known as drilling down on the data. Conversely, viewing information from the detailed level up to a higher level is known as rolling up on the data. For example, suppose there is a data warehouse containing citizen data for the whole country. Now a query analysis to find out the percentage of people areawise in different occupations can be carried out at city/village level, district level (each district comprising of several cities and villages), state level (each state comprising o several districts), and country level (the country comprising of several states). A user may carry out drill down analysis by starting the analysis at state level and gradually extracting this information for individual districts and cities/villages of interest. On the other hand, a user may carry out roll up analysis by starting the analysis at city/village level and gradually extracting aggregated information for districts, states, and the country. ROLAP and MOLAP We saw that OLAP systems are designed to support effective decision-making. The two popular architectures for OLAP systems are: 1. MOLAP (Multidimensional OLAP) and 2. ROLAP (Relational OLAP) The analytical ability and the interface of these two architectures are the same, what differs is the way data is physically stored. In MOLAP, data is stored in a
proprietary multidimensional database (MDDB) to be viewed in a multidimensional manner. On the other hand, in ROLAP, data is stored in a conventional relational database and query languages (like SQL) are used to perform analysis on the data in the data warehouse. The relative advantages and limitations of the two architectures are as follows: 1. Pre-identified queries are easier and faster to process with MOLAP than with ROLAP because the multiple dimensions of the multidimensional database can be properly structured to handle such queries efficiently. 2. Since MOLAP architecture stores the same data in the form of multiple dimensions, the storage space requirement is more in case of MOLAP than ROLAP. Hence MOLAP architecture has a limit on the physical database size it can handle. ROLAP architecture is scalable up to several terabytes. 3. MOLAP architecture can handle a limited number of dimensions. ROLAP architecture does not have any limit on the number of dimensions it can handle. 4. A proprietary multidimensional database is essential for MOLAP. ROLAP can be implemented using conventional relational database technology. Star and Snowflake Schemas The two commonly used schema models for designing data warehouses are: 1. The star schema, and 2. The snowflake schema In the star schema model, the data warehouse consists of one fact (major) table and many dimension (minor) tables. The fact table stores the primary keys for all dimension tables. A simple example of a star schema is shown in Figure 16.17. Here sales is a fact table, whereas customer, product, time, and store are dimension tables. The design appears like a star, with the fact table in the center, and dimension tables connected to it by radial lines. The main advantages of star schema are that it is easy to define, reduces the number of physical joins, and provides simple metadata. On the other hand, in the snowflake schema model, the data warehouse consists of normalized dimension tables, by attribute level with each smaller dimension table pointing to an appropriate aggregated fact table. Each dimension table has one key for each level of dimension's hierarchy. The lowest key, for example, the store key of store dimension will join store fact table, which will also have product and time keys. An example of a snowflake schema is shown in Figure 16.18. A practical way of designing a snowflake schema is to start with a star schema and create snowflakes with queries. The main advantage of snowflake schema is that it provides very good performance for queries involving aggregation.
DATA MINING The term 'mining' literally means the operations involved in digging for hidden treasures. Similarly, 'data mining' is used for the operations involved in digging out critical information from within an organization’s stored data for better decision support. It is the process of identifying interesting and useful patterns from large databases or data warehouses. As shown in Figure 16.14, data mining is the core part of the KDD process. Formally, data mining is defined as "the nontrivial process of extracting implicit, previously unknown, and potentially useful information from data". The meaning of the key terms used in this definition is as follows: 1. Nontrivial process. The term 'process' here means that data mining involves many steps including data preparation, search for patterns, knowledge evaluation, and refinement - all repeated in multiple iterations. This process is 'nontrivial' in the sense that it goes beyond simple search for quantities, and involves search for structure, models, and patterns. 2. Implicit information. This means that the information extracted as the result of data mining is from within an organization's stored data. 3. Previously unknown information. This means that the structures, models, or patterns extracted as the result of data mining are a new knowledge obtained which was previously unknown in the organization (hidden in the organization's data). 4. Potentially useful information. This means that the information extracted as the result of data mining is valuable and useful for the intended user or task (for better decision support). For being useful, it should also be understandable (if not immediately, then after some post processing). Classification of Data Mining Algorithms Many types of patterns normally exist in a large database. Different types of algorithms are needed to extract different kinds of patterns. Some of the most popular classes of algorithms used for data mining are: 1. Classification, 2. Clustering, 3. Association, and 4. Trend analysis They are briefly described below. Classification
The objective of classification algorithms for data mining is to classify the objects of a database into one of several predefined categorical classes. A classification algorithm is designed to determine the class of an object based on its attributes. In this method, a set of objects is first taken as the training set, and every object in this set is represented by a vector of attributes along with its class. Then a classification function or model is constructed by analyzing the relationship between the attributes and the classes of the objects in the training set. This classification function or model is then applied to all objects in a database to classify them into the already defined classes. This helps the user in having a better understanding of the classes of the objects in the database. A simple application of classification algorithm may be to classify the patient records of a patient database into different classes of diseases. The personal and clinical data of a patient's record may be used as the attributes for classification. The classified information may be used by the hospital management personnel to make decisions regarding, how many doctors and other staff should be employed with certain speciality and skills, how many beds should be provided in different sections of the hospital, what types of medicines and in how much quantity should be stocked, and so on. Clustering The objective of clustering algorithms for data mining is to automatically partition the data space of a database into a set of clusters, to which each of the objects in the database are assigned, either deterministically or probabilistically. A clustering algorithm is designed to identify all sets of similar objects in a database, in some optimal fashion. The objects are so clustered that the intra-cluster (objects belonging to the same cluster) similarities are maximized, and the inter-cluster (objects belonging to different clusters) similarities are minimized. This is done based on some criteria defined on the attributes of the objects. Once the clusters are decided, the common features for objects in a cluster are summarized to form the cluster description. It is important to understand the difference between classification and clustering algorithms. In the case of classification, the objects of a database are classified into predefined classes whose number is known and fixed before the classification algorithm is applied to the database. However, in case of clustering, the objects of a database are clustered into clusters whose number is not known before the clustering algorithm is applied to the database. Thus, if the same classification algorithm is applied to two different databases, both will result in the same predefined number of classes (say n, some of which may be null if no object mans to them). On the other hand, if the same clustering algorithm is applied to two different databases, both may result in different, previously unknown number of
clusters (say m and ri). Also note that the common features of objects in a class are pre-defined in the classification algorithm, but the common features of objects in a cluster are summarized after all the objects have been put in the cluster. A simple application of clustering algorithm may be to cluster the customers of a bank from the bank's customer database into several groups based on similarities of their age, sex, income, and residence. Once clustering is done, the bank management can examine the summarized characteristics of each subgroup, extract significant statistics, and use them to understand its customers better and thus provide more suitable products and customized services. Clustering -algorithms have also been successfully applied in the area of text mining for problems such as document indexing and topic identification in the research setting. Association The objective of association algorithms for data mining is to automatically discover association rules hidden in a database. An association rule reveals the associative relationships among objects. For example, an association rule may reveal that "If some object X is part of a transaction, then for some percent of such transactions, object Y is also part of the transaction." Thus association algorithms help in the discovery of togetherness, or the connection of objects. They are useful in solving problems where it is important to understand the extent to which the presence of some variables imply the presence of other variables and the prevalence of this pattern in the entire database under examination. An interesting and very useful application of association algorithm is marketbasket analysis, in which database of sales transactions are examined to extract patterns that identify what items sell together, what items sell better when relocated to new areas, and what product groupings improve department sales. For example, a retail store may discover that people tend to buy soft drinks and potato chips together. Store personnel then place the potato chips near the soft drinks to promote the sale of both. They may even discount one to entice buying the other, since these buyers now will be saving some money. Another useful application of association algorithm is the analysis of Internet traffic on a web site to better understand where the real demand is, what pages are being looked collectively, and so on. Service providers can use this information to better organize their web pages. Trend Analysis
Several large databases have time series data (records accumulated over time) with timestamp. For example company's sales data over several years/months, a customer's credit card transactions over a period of time, a the fluctuations in stock prices over a period of time, are all time series data. Such data can be easily analyzed find certain historical trends. The objective of trend analysis algorithms for data mining is to discover the patterns and regularities in the data evolutions along with the dimension of time. Such patterns have often been found to be very useful in making decisions for larger benefits in the future. A simple application of trend analysis algorithm may be to analyze the historical trend of fluctuations in the stock prices of various companies, to predict future behavior. Such analysis results can be effectively used by investors in the stock market for making investment decisions for better profitability. Data Mining Techniques Data mining adopts several techniques drawn from many different areas for implementing the various types of algorithms-discussed above. Some of the most popular techniques used for data mining are: 1. Statistics, 2. Machine learning, 3. Database-oriented, 4. Neural networks, 5. Fuzzy sets, and 6. Visual exploration They are briefly described below. Most data mining systems employ multiple of these techniques to deal with different kinds of data, different data mining tasks, and different application areas. Statistics Statistical methods are very useful in data mining. Usually, statistical models are built from a set of training data. An optimal model, based on a defined statistical measure, is searched among the hypothesis space. Rules, patterns, and regularities are then drawn from the model. Some of the commonly used statistical methods used in data mining are: 1. Regression analysis. It maps a set of attributes of objects to an output variable. 2. Correlation analysis. It studies the correspondence of variables to each other, such as the x2. 3. Cluster analysis. It finds groups from a set of objects based on distance measure.
4. Bayesian network. It represents the causal relationships (what object/event is caused by which other objects/events) among the variables in the form of a directed graph, which is computed by using the Bayesian probability theorem. Machine Learning Like statistical methods, machine learning methods search for the model that is closest (best matches) the data being tested. However, unlike statistical methods, the searching space is a cognitive space of n attributes instead of a vector space of n dimensions. Besides that, most machine learning methods use heuristics in the search process. Some of the commonly used machine learning methods used in data mining are: 1. Decision tree. Decision trees are a way of representing a series of rules for the classification of a dataset and consist of nodes and branches. They are built using a training set of data and are used to classify objects in a dataset. An object's class is determined by the use of a decision tree by following the path from the root to a leaf node. It chooses the branches according to the attribute values of the object. A simple decision tree that may be used for classifying the employees of an organization into groups based on their age, sex, and salary is shown in Figure 16.19. Decision trees have proved to be an attractive tool for data mining because they are simple to implement and their intuitive representation makes the resulting classification model easy to understand. 2. Inductive concept learning. It derives a concise, logical description of a concept from a set of examples. 3. Conceptual clustering. It finds groups or clusters in a set of objects based on conceptual closeness among objects. Database-oriented Database-oriented methods use data modeling or database specific heuristics to exploit the characteristics of data in hand. Some of the commonly used databaseoriented methods used in data mining are: 1. Attribute-oriented induction. It generalizes primitive, low-level data into high-level concepts by using conceptual hierarchies. 2. Iterative database scanning. It is employed to search for frequent item sets in a transactional database. The association rules are then derived from these frequent item sets. 3. Attribute focusing. It looks for patterns with unusual probabilities by adding attributes selectively into the patterns. Neural Networks
A neural network is a set of interlinked nodes called neurons. A neuron is a simple device that computes function of its inputs. The inputs can be outputs of other neurons or attribute values of an object. By adjusting the connection and the functional parameters of the neurons, a neural network can be trained to model the relationship between a set of input attributes and an output attribute. Such models attempt to mimic brain activity by adapting the weights of the interconnections among the neurons in the network allowing learning and memory creation to take place. The neural network method is ideal for prediction and classification in situations where there are a good deal of historical data available for training. Fuzzy Sets A fuzzy set is a set whose membership is fuzzy (based on rough match of attribute values instead of exact matches). In conventional classification schemes, we look for exact matches of attribute values (either true or totally false), but in systems based on fuzzy sets, truth values can lie anywhere on the 0.0 to 1.0 interval of real numbers. Using logical and fuzzy operators, such as AND, OR, NOT, VERY, and SOMEWHAT, the system can make fuzzy decisions. For example, IF (B OR C) THEN D can be stated in a system based on fuzzy sets as IF (SOMEWHAT B OR VERY C) THEN D Systems based on fuzzy sets can be used to form a group of fuzzy sets for use, in say, classification and clustering. Visual exploration In this method, data is transformed into visual objects (such as dots, lines, and areas) and displayed in a two or three dimensional space. Users then interactively explore the interesting spots and useful patterns by usual examination. Data Warehousing and Data Mining It is important to know the difference between data warehousing and data mining. As can be seen from Figure 16.14, the two are related activities and can reinforce each other. Usually, they are used in conjunction with each other. A data warehouse sets the stage for effective data mining. Data mining can be done where there is no data warehouse, but the data warehouse greatly improves the chances of success in data mining. Data mining is also different from OLAP, in that the latter is used to verify more complex human-generated hypotheses involving multiple dimensions, produces information summaries, and does not look for patterns. Thus OLAP is knowledge
verification, while data mining is knowledge discovery. Furthermore, OLAP is multidimensional analysis and data mining is multi-variable, single-dimensional analysis. For example, while OLAP answers the questions of what is the average credit granted to customers who pay their monthly bills in full versus those who choose to take financing, data mining gives the attributes associated with customers who will take up financing. OTHER RELATED CONCEPTS Database Security The data stored in a database is often confidential and should not be accessed by unauthorized individuals. To prevent data stored in a database from being accessed by unauthorized individuals, database management systems commonly support the following security features: 1. From the point of view of access control, a database management system normally supports two types of database users - database administrator (having administrative access) and database user (having normal user access). A database administrator has privileges to add new users to the database, delete one or more users of a database, decide and enforce the various access rights of different users, make changes to the schema, and so on. On the other hand, a normal user can only view, modify, delete or add records; can enter data; can search for desired information; and can create and generate reports. 2. All database management systems have password-based user authentication mechanism. That is, all database users must enter their valid password to successfully authenticate themselves to the database system in order to obtain access to all or part of the data in the database. 3. Many database management systems support a program-level of security that precludes unauthorized users from using particular programs that access the database. 4. Many database management systems support record-level security that prevents unauthorized users from accessing or updating certain records, such as employee records of top-level managers in an employee database. 5. Many database management systems even support field-level security that prevents unauthorized users from accessing or updating sensitive information in certain fields, such as salary. 6. Many database management systems support read-only access that prevents unauthorized users from modifying certain information (they can only view the information). 7. All database management systems maintain a log of database accesses so that in case of any security breach, the log records can be analyzed to detect the possible cause of the problem. Client-Server Database Systems
With the proliferation of computer networks, it became possible to access the data stored in a database from any system connected to the same network on which the system having the database was located. This gave birth to the client-server model of database systems. As shown in Figure 16.20, in this model the database is stored at and managed by a computer system called the database server. It runs all important modules of the database management system that deal with the maintenance and processing of data (called server programs). On the other hand, systems on the same network that are primarily meant for accessing data from the database are called database clients. They serve as front-end systems to the database and run those modules (called client programs) of the database management system that facilitate specification of access requests. When a database client needs some information from the database, it makes a request to the database server. The request travels on the network from the database client to the database server, and the database server retrieves the requested information from the database and sends only that information back to the client. Obviously, there can be many database clients simultaneously connected to the same database server and all of them can share the same database. The client-server database system enjoys the following advantages over conventional database systems: 1. Efficient utilization of computing resources. Separate databases are not required for every user. Hence only the database server needs to have high processing power and large storage. The database clients can be low-end personal computers having low processing power and low storage capacity. 2. Since only one copy of the database is maintained for all the clients, it is easier to maintain the integrity (consistency) of the data. That is, all database updates need to be carried out only at one place. 3. Since only one copy of the database is maintained for all the clients, it is also easier to ensure greater security of the data. 4. The system administration and maintenance cost is also lower because a good system and database administrator is required only where the database server is. With the popularity of the Internet, the4concept of database clients has now been extended to web clients. A web client is a computer system connected to the Internet that runs a web browser. The web browser enables a web client to dynamically connect to a database server, download necessary client software for becoming a database client, and make requests for information to the database server. The web-based database clients have proved to be better than the conventional database clients because the deployment of the client software to hundreds of database client machines was causing unacceptable effort. That is, seamless upgrades of client software to new versions was either not possible or was causing high maintenance cost.
Distributed Database Systems Another development in the database area that was the result of the proliferation of computer networks was that of distributed database systems. A distributed database system seamlessly integrates multiple databases located on different computers. These computers may be located in the same building or may be geographically far apart (such as in different cities). Note that unlike a clientserver database system in which there is only one database, a distributed database system has multiple databases integrated together to serve as a single database. The main motivation behind using a distributed database system is to store data in the most convenient places within a computer network, depending on data entry requirements, storage capacity availability, processing load requirements, data security requirements, and so on. For example, as shown in Figure 16.21, a distributed database system might integrate employee databases of different departments of an organization that are located in different cities. Each department's database stores the records of the employees working in that department. A distributed database management system functions in such a manner that the multiple distributed databases managed by it appear as a single centralized database to the end users. That is. a user making a request would not be concerned about the source of the data or how they are retrieved. For example, in the distributed database system of Figure 16.21, a user can simply make a single request for a sorted list of all female employees in the organization who are more than 40 years old. The distributed database management system takes care of sending this request to all the four databases, then accumulating the results, sorting it, and presenting it to the user. All this is done internally without the knowledge of the user making the request. Active Database Systems Traditional database management systems are passive in the sense that commands are executed by the database (such as, query, update, delete) as and when requested by the user or application program. However, some situations cannot be effectively modeled by this pattern. For example, consider an inventory database of a department store, which stores the current quantity of each item along with its price and registered vendor's details. The database is directly connected to the point-of-sale terminals of the store and as the items are sold, the inventory in the database keeps getting updated. In order to ensure that an item never runs out of stock in the store, it is required that an order for the item is placed as soon as its quantity in the inventory falls below a threshold value. To do this in a passive database system, a polling mechanism can be implemented that periodically
checks the quantity of every item in the inventory and prints a list of those whose quantity is less than its threshold value. New orders can then be placed for the items in this list. The main difficulty in implementing this mechanism is how to decide the most appropriate polling frequency. If it is too high, there is a cost penalty. On the other hand, if it is too low, the reaction may be too late to meet the requirement of no item running out of stock. The best way to handle such an application would be to make the system automatically place an order with the registered vendor as soon as the current quantity of an item goes below its threshold value. This is exactly what an active database system does. For instance, in case of our example, the system can be designed to automatically invoke an application when the quantity of an item goes below its threshold value. This application prepares a purchase order for the item and sends this order to the registered vendor on e-mail. An active database system supports applications of the type mentioned above by moving the reactive behavior from the application (or polling mechanism) into the database management system. Active databases are thus able to monitor and react to specific circumstances of relevance to an application. An active database system must provide a knowledge model (that is, a description mechanism) and an execution model (that is, a runtime strategy) for supporting such applications. For this, an active database system uses rules that have up to three components - an event, a condition, and an action. The event part of a rule describes a happening to which the rule may be able to respond. The condition part of the rule examines the context in which the event has taken place. The action describes the task to be carried out by the rule if the relevant event has taken place and the condition has evaluated to true. Such a rule is known as eventcondition-action or ECA-rule. In some cases, the event or the condition may be ether missing or implicit. For instance, in the application discussed above, there is no condition. Hence it is an example of event-action rule. The event is "quantity below threshold value" and the action is "invoke the application for placing order". Points to Remember 1. Data processing is defined as a series of actions or operations that converts data into useful information. Thus, data is the raw material of information and just as raw materials are transformed into finished products by a manufacturing process, raw data is transformed into information by data processing. 2. A data processing system includes the resources such as people, procedures, and devices that are used to accomplish the processing of data for producing desirable output. 3. In data processing, storage of data is often conceptualized as a data storage hierarchy, which normally consists of bit, character, field, record, file and database.
4. Multiple related bits are combined to form a character, multiple related characters are combined to form a field, multiple related fields are combined to form a record, multiple related records are combined to form a file, and multiple related files are integrated to form a database. 5. The two standard methods used in data processing for organizing data are fileoriented approach and database-oriented approach. 6. In the file-oriented approach of organizing data, an application's data is organized into one or more files and the application program processes the data stored in these files to generate the desired output. A set of programs is provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their files. All these programs together form a File Management System (FMS). 7. In the database-oriented approach of organizing data, data from multiple related files is integrated together in the form of a database, which has the following properties: • It provides greater query flexibility. • It reduces data redundancy. • It solves data integrity (inconsistency) problem. • It makes data independent of the application programs. That is, fields can be added, changed, and deleted from the database without affecting existing programs. • It also includes data security features at database level, record level, and even at field level to provide greater flexibility of restricted data access. A set of programs is provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their data in a database. All these programs together form a Database Management System (DBMS). 8. File organization deals with the physical organization of the records of a file for the convenience of storage and retrieval of data records. The three commonly used file organizations in business data processing applications are sequential, direct/random, and indexed sequential. 9. File utilities consist of routines, which perform a variety of generalized operations on data files. The operations performed by some of the commonly used file utilities are sorting, searching, merging, copying, printing and maintenance. 10. A database model defines the manner in which the various files of a database are linked together. The four commonly used database medels are hierarchical, network, relational, and object-oriented. 11. The four major components of a database management system are data definition language (DDL), data manipulation language (DML), query language and report generator. 12. The data definition language (DDL) is used to define the structure of a database. The database structure definition is commonly known as database schema.
13. The data manipulation language (DML) includes all the commands that enable the users to enter and manipulate the data. 14. The query language enables users to define their requirements for extracting the desired information from the database in the form of queries. 15. The report generator enables the users of a database to design the layout of a report so that it can be presented in the desired format. 16. A multimedia database is a collection of multimedia data items, such as text, graphics, animation, audio, and video. A multimedia database management system is a set of programs to define, create, store, access manage, and query the multimedia database. 17. Special features needed in a multimedia database management system for handling multimedia data include • Support for multimedia data storage and retrieval • Support for multimedia data types • Support for multimedia data relationships • Support for multimedia data modeling • Support for multimedia data manipulation • Support for multimedia data query and • Support for multimedia user interface 18. The techniques and tools meant for extraction of knowledge from large datasets are the subject of the field called knowledge discovery in databases (KDD). 19. The KDD process typically consists of the following steps: • Data selection • Data preprocessing • Data transformation • Data mining and • Data interpretation/evaluation 20. Data warehousing is the process of creating a repository of integrated data for the purpose of decision support and analysis. A data warehouse is the resulting repository of integrated data. It is a subject-oriented, integrated, time-variant, and non-volatile collection of data in support of management's decision-making process. 21. Data mining is the process of identifying interesting and useful patterns from large databases or data warehouses. It is defined as the nontrivial process of extracting implicit, previously unknown, and potentially useful information from data. 22. A distributed database system seamlessly Integrates multiple databases located on different computers. A distributed database management system functions in such a manner that the multiple distributed databases managed by it appear as a single centralized database to the end users. 23. An active database system is capable of monitoring and reacting to specific circumstances of relevance to an application.
Questions 1. What is the difference between data and information? 2. What is meant by data processing? 3. What is a data processing system? 4. In data processing, storage of data is often conceptualized as a data storage hierarchy. Describe the various levels of this hierarchy. 5. Give an example to illustrate the relationship between a character, a field, a record, and a file. 6. What is a database? How is it different from a file? 7. What are the two standard methods used in data processing systems for organizing data? Explain their relative advantages and disadvantages. 8. Describe three drawbacks that occur in traditional information processing when separate, unrelated files are used. 9. What is the difference between a master file and a transaction file? 10. What is a key field? How is it useful? 11. What is data redundancy? Explain with an example. Now explain how the database-oriented approach of data organization helps in reducing data redundancy as compared to the file-oriented approach. 12. What is data integrity problem? Explain how the database-oriented approach of data organization helps in solving this problem. 13. What is a file management system? 14. Describe the various types of files commonly supported in a typical file management system. 15. What is the difference between an output file and a report file? 16. What is a backup file? Why is regular creation of backup files important? 17. What is meant by file organization? What are the three commonly used file organizations in business data processing? How do we decide to use a particular file organization for an application? 18. What is a sequential file? What is the principal storage medium used for storing sequential files? Give examples of few applications that will benefit from the use of sequential files. 19. What is activity ratio in file processing? What type of file organization is suitable for (a) an application that has high activity ratio, (b) an application that has low activity ratio? Give an example of each type c application. 20. How is a sequential file organized? How are records in a sequential file accessed? How are these record processed? 21. What is a direct file? What is the principal storage medium used for storing direct files? Give examples of few applications that will benefit from the use of direct files. 22. How is a record stored in a direct file? How is it retrieved and processed?
23. Explain how are the records in a direct file physically organized so that it becomes possible to directly access a particular record, given its key field value. 24. What is hashing? How is it useful in direct file organization? What is a collision? How can it be taken can care of? 25. If the activity ratio is high, the use of direct files proves to be very inefficient as compared to sequential files. Explain why. 26. Explain the basic principle used in indexed sequential file organization. Explain the roles of data file and index file in indexed sequential file organization. 27. How are records stored in an indexed sequential file? How are they retrieved and processed? 28. Explain how can indexed sequential files be used quite efficiently for both sequential and direct access processing. 29. Discuss the advantages and limitations of the sequential, direct, and indexed sequential file organizations. 30. What are file utilities? 31. Differentiate between the process of sorting and merging of files. 32. Explain the use of the following file utilities: copying, searching, printing, and maintenance. 33. What is a database management system? 34. Give examples of how a company and an individual might use a DBMS. 35. What is a database model? Name the four commonly used database models. Describe any two of these models. 36. Write short notes on: (a) Hierarchical database model (c) Relational database model (b) Network database model (d) Object-oriented database model 37. Explain the relational database model with the help of an example. List out the main advantages of the relational database model as compared to the hierarchical and network database models. 38. The collection of records in a database may have different sets of fields. Discuss this statement. Give an example to illustrate this. 39. If new type of data is to be added to an existing relational database, it is not necessary to redesign the database afresh. Discuss this statement. Give an example to illustrate this. 40. List out some of the key features found in the object-oriented database model that are absent in conventional database models. 41. Why is the object-oriented database model preferred for database applications such as CAD, CAM, CAE, CASE, expert systems and multimedia systems? 42. Explain the object-oriented database model with the help of an example. 43. What are the four major components of a database management system (DBMS)? Describe the roles of each of these components in the overall design and usage of a database.
44. What is a data definition language (DDL)? What is its role in the overall design and/or usage of a database? 45. What is a data manipulation language (DML)? What is its role in the overall design and/or usage of a database? 46. What is a query language? What is its role in the overall design and/or usage of a database? 47. What is the difference between the data manipulation language and the query language with respect to what each enables the user to do? 48. What is a report generator? What is its role in the overall design and/or usage of a database? 49. A report generator can be instructed to perform arithmetic operations on the data found in numeric fields to make a report more meaningful and useful. Discuss this statement. Give an example to illustrate this. 50. What is a database schema? What all does it typically include about a database? 51. What are the jobs and responsibilities of a database administrator? 52. What is a data dictionary? How is it created/updated? 53. What is SQL? How is it useful? 54. List out in proper sequence the three basic steps involved in creating a database? Briefly describe these steps. 55. What are forms? List out some of the features that may be incorporated into a form to facilitate easier data entry. 56. What is a filter? Explain with an example. 57. List out the three commonly supported features in modern database systems that may be used for searching for desired information from a database. 58. Explain with an example how "Find" command can be used for making a database query? What are the limitations of "Find" command? 59. Explain with an example how a query language can be used for making a database query? What are the advantages of using a query language against "Find" command? 60. What is query by example (QBE)? How does it make the job of querying a database simpler? 61. A DBMS derives much of its power from the ease with which its users can search for the desired information from the large volume of data stored in a database. Discuss this statement. 62. What is a report? How is it created and used? Give an example of a sample report. 63. What is a multimedia database system? What are the typical requirements of multimedia database systems? 64. Discuss how the following requirements of multimedia database systems are different from that of conventional database systems: (a) Storage devices (c) Data manipulation (b) Data modeling (d) Database querying
65. What is content-based retrieval/querying (CBR/CBQ)? In what type of databases is this feature required an why? 66. What is derived data with reference to a multimedia database system? Why is it used? 67. Explain with an example for each of the following types of relationships that need to be taken care of in multimedia database systems: (a) Spatial relationship (b) Temporal relationship (c) Synchronization relationship 68. Describe any two database query methods that are typically used in multimedia database systems. 69. Write short notes on: (a) Database security (b) Active database systems (c) Multimedia database systems (d) Client-server database systems (e) Distributed database systems 70. Write short notes on: (a) Storage requirements of multimedia databases (b) Query features in multimedia database systems (c) Relationships among data in multimedia database systems (d) Data manipulation features in multimedia database systems 71. Write short notes on: (a) Data storage hierarchy (b) Output versus report files (c) Backup files (d) Master versus transaction files (e) File-oriented versus database-oriented approaches of organizing data 72. Write short notes on: (a) Sequential file organization (b) Direct file organization (c) Indexed sequential file organization 73. Write short notes on: (a) Sorting (d) Copying (b) Searching (e) Printing (c) Merging (f) Maintenance 74. Write short notes on: (a) Database models (c) Components of a DBMS (b) Database schema (d) Query features in database systems 75. What is meant by knowledge discovery in databases (KDD)? What is the main motivation behind KDD? 76. Describe the KDD process.
77. Differentiate among the terms knowledge discovery in databases (KDD), data warehousing, and data mining. 78. Explain the following terms with respect to the KDD process: (a) Data selection (c) Data transformation (b) Data cleaning (d) Data mining 79. The KDD process is interactive and iterative. Discuss. 80. What is a data warehouse? Explain its key features. 81. A data warehouse is defined as a subject-oriented, integrated, time-variant, and non-volatile collection of data. Explain the meaning of the key terms used in this definition. 82. When does an organization typically need to build a data warehouse? 83. Describe the process of building a data warehouse. 84. Differentiate between OLTP and OLAP. 85. What is a data mart? How is it different from a data warehouse? 86. What is metadata? How is it useful? 87. Differentiate between drill down and roll up analyses with reference to data warehouses. 88. Differentiate between ROLAP and MOLAP. List out their relative advantages and limitations. 89. Differentiate between star and snowflake schema models. Give an example of each. 90. Write short notes on: (a) Data warehousing (b) Data mart 91. Write short notes on: (a) OLTP versus OLAP (b) ROLAP versus MOLAP (c) Metadata (d) Data mining (c) Star versus snowflake schema (d) Drill down versus roll up analysis 92. What is data mining? How is it useful? Explain with an example. 93. Data mining is defined as the nontrivial process of extracting implicit, previously unknown, and potentially useful information from data. Explain the meaning of the key terms used in this definition. 94. In data mining, different types of algorithms are needed to extract different kinds of patterns. Briefly describe some of the most popular classes of algorithms used for data mining. 95. Give an example each to explain the concept of classification, clustering, association, and trend analysis used in data mining. 96. Briefly describe the following techniques used in data mining: (a) Statistics (d) Fuzzy sets (b) Machine learning (e) Neural networks
(c) Database-oriented (f) Visual exploration 97. Differentiate between data warehousing and data mining. 98. Explain how is data mining different from OLAP. 99. Write short notes on: (a) Data mining techniques (b) Data mining versus OLAP (c) Classes of data mining algorithms (d) Data mining and data warehousing 100. List out some of the commonly supported security features in a database management system. 101. What is a client-server database system? Explain the roles of the database server and database clients in such a system. List out some of the advantages of a client-server database system over conventional database systems. 102. What is a web client in a client-server database system? How is it better than a conventional database client? 103. What is a distributed database system? What is the main motivation behind using such a system? 104. What is an active database system? Explain how an active database system monitors and reacts to specific circumstances of relevance to an applicati Chapter 17 Data Communications and Computer Networks It is not the processing power of computers, but the power of communication among computers that has revolutionized the information age. The marriage of computing and data communication technologies (means and methods by which data is transferred from one location to another) is one of the most exciting developments in today's information age. This development has given birth to computer networks. A computer network is a network of geographically distributed multiple computers connected in such a way that meaningful transmission and exchange of information become possible among them. Sharing of information, sharing of resources (both hardware and software), and sharing of processing load are some of the major objectives of a computer network. The goal of this chapter is to introduce the various aspects of data communications technology and to discuss how this technology is combined with computer technology to form computer networks. Following are some of the important concepts, which you will learn from this chapter: 1. Basic elements of a communication system. 2. The techniques, channels and devices used to transmit data between distant locations. 3. The various types of computer networks.
4. Communication protocols, which are rules and procedure for establishing and controlling the transmission of data from one computer to another. 5. Characteristics and advantages of distributed data processing. Communication is the process of transferring messages from one point to another. As shown in Figure 17.1, the three basic elements of any communication process are: 1. A sender (source), which creates the message to be transmitted. 2. A medium, which carries the message. 3. A receive (sink), which receives the message. For example, when you speak to your friend on the telephone, you are the sender. The telephone line through which your voice is transmitted is the medium and your friend is the receiver. This is a simple example of voice communication. The same concept holds good for data communication also. Data communication is the function of transporting data from one point to another. In this case, the sender and the receiver are normally machines, in particular, computer devices (computers, terminals, peripheral devices like line printers, plotters, disks, etc.) and the transmission medium may be telephone lines, microwave links, satellite links, etc. However, the messages that are transmitted are data, not voice conversations. Thus, the electronic systems that transfer data from one point to another are called data communication systems. Unlike computers that process and rearrange data, data communication systems transmit data from one point to another without any change. DATA TRANSMISSION MODES There are three ways, or modes, of transmitting data from one point to another. As shown in Figure 17.2, the; are simplex, half-duplex, and full-duplex. Simplex If transmission is simplex, communication can take place in only one direction. Devices connected to such a circuit are either a send-only or a receive-only device. For example, a data collection terminal on a factory floor (send only) or a line printer (receive only). At first thought, that might appear adequate for many types of applications in which flow of information is unidirectional. However, in almost all data processing applications, communication in both directions is required. Even for a "one-way" flow of information from a terminal to a computer, the system will be designed to allow the computer to signal the terminal that data has been received. Without this capability, the remote user might enter data and never know that it was not received by the computer (due to some problem somewhere). Hence, simplex circuits are seldom used because a return path is generally needed to send acknowledgement, control, or error signals.
Half-Duplex A half-duplex system can transmit data in both directions, but only in one direction at a time. Thus, a half-duplex line can alternately send and receive data. It requires two wires. This is the most common type of transmission for voice communication because only one person is supposed to speak at a time. It is also used to connect a terminal with a computer. The terminal might transmit data and then the computer responds with an acknowledgement. Full-Duplex In a half-duplex system, the line must be "turned-around" each time the direction is reversed. This involves a special switching circuit and requires a small amount of time (approximately 150 milliseconds). With high-speed capabilities of the computer, this turn-around time is unacceptable in many instances. Also, some applications require simultaneous transmission in both directions. In such cases, a full-duplex system is used that allows information to flow simultaneously in both directions on the transmission path. Use of a full-duplex line improves efficiency as the line turn-around time required in a half-duplex arrangement is eliminated. It requires four wires. DATA TRANSMISSION SPEED A term used to describe the data-handling capacity of a communication system is bandwidth. Bandwidth is the range of frequencies that is available for the transmission of data. A narrow range of frequencies in a communication system is analogous to a narrow road. The flow of information in such a system - its data transfer rate - is limited, just as is the flow of traffic in a narrow road. Wider bandwidths permit more rapid information flow. Thus, the wider the bandwidth of a communication system, the more data it can transmit in a given period of time. When dealing with computer input/output devices, we think in terms of characters per second. However, in case of data transmission, we talk in terms of bits per second. The ASCII code uses 7 data bits per character plus a parity bit. For data communication, additional bits are added to control the process. Although the number of bits depends upon the communication system used, commonly encountered systems use a total of either 10 or 11 bits per character. Hence a terminal having a speed of 30 characters per second would probably be used with a communication system, which transmits at the rate of 300 bits per second. The communication data transfer rate is measured in a unit called baud. In general usage, baud is identical to bits per second. For instance, a rate of 300 baud is 300 bits per second. Therefore the 30 characters per second terminal would be said to operate at 300 baud. However, technically baud refers to the number of signal
(state) changes per second. Thus, using more sophisticated coding techniques, 1 baud can represent 2 or even 3 bits per second. But, with most communication systems, 1 baud represents only one signal change per second and thus is equivalent to 1 bit per second. Depending on their transmission speeds, communication channels (paths) are grouped into three basic categories -narrowband, voiceband, and broadband. Narrowband Narrowband or subvoice grade channels range in speed from 45 to 300 baud. They are used for handling low data volumes and are adequate for low-speed devices. They are used mainly for telegraph lines and low speed terminals. Voiceband Voiceband channels handle moderate data volumes and can transmit data at speeds up to 9600 baud. They are so called because their major application is for ordinary telephone voice communication. They are also used for data transmission from slow I/O devices to CPU or from CPU to slow I/O devices. Moreover, most remote terminals are connected to computers through voiceband channels. Broadband Broadband or wideband-channels are used when large volumes of data are to be transmitted at high speed. These systems provide data transmission rates of 1 million baud or more. A company might use a broadband facility for high-speed computer-to-computer communication or for simultaneous transmission of data to several different devices. As might be expected, the cost of the service increases with speed. Hence a thorough analysis of the business needs and associated costs is necessary in making a proper choice. DATA TRANSMISSION MEDIA The sender-medium-receiver concept has actually been with us for a very long time. For example, shouting to another person does in fact involve voice transmission over a distance via the medium air, which carries the sound wave. Needless to say, the use of telephone lines as a transmission medium considerably enhances the possible distance. Like telephone lines, there are several types of physical channels (communication media) through which data can be transmitted from one point to another. Some of the most common data transmission medium are briefly described below.
Twisted-Pair Wire A twisted-pair wire consists of two bunches of thin copper wires, each bunch enclosed separately in a plastic insulation, then twisted around each other to reduce interference by adjacent wires (see Figure 17.3). It is also called unshielded twisted-pair (UTP) cable because other than the plastic coating around the two individual bunches of copper wires, nothing shields it from outside interference. UTP cables are commonly used in local telephone communication and short distance (up to about 1 km) digital data transmission. They are normally used to connect terminals to the main computer if they are placed at a short distance from the main computer! Data transmission speed of up to 9600 bits per second (or 9600 baud) can be achieved if the distance is not more than 100 meters. However, for longer distance data transmission, local telephone lines are used. In this case, typical speed of digital signal transmission is 1200 bits per second. UTP cables are an inexpensive medium of data transmission. They are easy to install and use. However, their use is limited because they easily pick up noise signals, which results in high error rates when the line length extends beyond 100 meters. Coaxial Cable Coaxial cables are groups of specially wrapped and insulated wire lines that are able to transmit data at high rates. As shown in Figure 17.4, they consist of a central copper wire surrounded by a PVC insulation over which a sleeve of copper mesh is placed. The metal sleeve is again shielded by an outer shield of thick PVC material. The signal is transmitted by the inner copper wire and is electrically shielded by the outer metal sleeve. Coaxial cables offer much higher bandwidths than UTP cables and are capable of transmitting digital signals at rates of 10 mega bits per second. They are extensively used in long distance telephone lines and as cables for cable TV. They are also used by telephone companies to transmit data. In many cases, several coaxial cables are packaged into a very large cable that can handle over 40,000 telephone calls simultaneously. Furthermore, coaxial cables have much higher noise immunity and can offer cleaner and crisper data transmission without distortion or loss of signal. Microwave System Another popular transmission media is microwave. This is a popular way of transmitting data since it does not I incur the expense of laying cables. Microwave systems use very high frequency radio signals to transmit data I through space. However, at microwave frequencies, the electromagnetic waves cannot bend or
pass obstacles like I hill. Hence it is necessary that microwaves transmission be in a line-of-sight. In other words, the transmitter and I receiver of a microwave system, which are mounted on very high towers, should be in a line-of-sight. This may I not be possible for very long distance transmission. Moreover, the signals become weaker after traveling a certain distance and require power amplification. In order to overcome the problem of line-of-sight and power amplification of weak signals, microwave systems use repeaters at intervals of about 25 to 30 kms in between the transmitting and receiving stations (see Figure 17.5). The first repeater is placed in line-of-sight of the transmitting station and the last repeater is placed in line of-sight of the receiving station. Two consecutive repeaters are also placed in line-of-sight of each other. The data signals are received, amplified, and retransmitted by each of these stations. Microwave systems permit data transmission rates of about 16 Giga (1 Giga = 10 9) bits per second. Furthermore, it can carry literally thousands of voice channels at the same time. The link can support 250,000 voice channels. The initial installation cost of microwave links being very high, they are mostly used to link big cities with heavy telephone traffic between them. Communications Satellite The main problem with microwave communication is that the curvature of the earth, mountains, and other structures often block the line-of-sight. Due to this reason several repeater stations are normally required for long distance transmission, which increases the cost of data transmission between two points. This problem is overcome by using satellites, which are relatively newer, and more promising data transmission media. A communication satellite is basically a microwave relay station placed in outer space. These satellites are launched either by rockets or by space shuttles and are precisely positioned 36,000 kms above the equator with an orbit speed that exactly matches the earth's rotation speed. Since a satellite is positioned in a geosynchronous orbit, it is stationary relative to earth and always stays over the same point on the ground. This allows a ground station to aim its antenna at a fixed point in the sky. Each satellite can receive and retransmit signals to slightly less than half of the earth's surface. Therefore, at least three satellites are needed in geosynchronous orbit to provide worldwide data transmission service. Hundreds of satellites are now in orbit to handle international and domestic data, voice, and video communications needs. The INSAT series of Indian satellites are positioned in such a way that they are accessible from any place in India.
As shown in Figure 17.6, in satellite communication, microwave signal at 6 GHz or 14 GHz (read as gigahertz = 10 Hz) is transmitted from a transmitter on earth to the satellite positioned in space. By the time this signal reaches the satellite, it becomes weak due to 36,000 kms travel. A transponder mounted on the satellite amplifies the weak signal and transmits it back to the earth at a frequency of 4 GHz or 1 lGHz. This signal is received at receiving station on the earth. It may be noted that the retransmission frequency is different from the transmission frequency in order to avoid the interference of the powerful retransmission signal with the weak incoming signal. A transponder can typically support 1200 voice channels each of 4800 bps or 400 digital channels each of 64Kbps data transmission rate. A satellite has many transponders. Therefore, a single satellite has enormous data communication capability. The use of 4 GHz band of frequencies for transmission and retransmission of microwave signals in a satellite communication system is called the C-band transmission. On the other hand, the use of 11 GHz to 14 GHz band of frequencies is called the Ku-band transmission. In order that a large number of users can benefit from satellite communication, it was realized that the size and cost of the receive-transmit earth stations has to be considerably reduced so that one can easily install a private earth station. To meet this requirement, smaller dish antennas having lower power transmitters and receivers have been built for installation at users' sites. These dish antenna systems are popularly known as VSATs (Very Small Aperture Terminals). The aperture refers to the diameter of the dish antenna, which is about 1 to 2 meters. Satellite communication systems have the following advantages: 1. A satellite, which is essentially a microwave relay station visible from any point of a very large area, can be used for transmission and reception of data between any two randomly chosen points in that area. 2. Data transmission costs are independent of the distance between two points as long as the two points are within the area covered by the satellite. 3. A satellite having many transponders has enormous data communication capability. 4. Error detection is trivial in a satellite communication system because a transmitting station can receive back its own transmission and check whether the satellite has transmitted the information correctly. If not, it can retransmit the data. A satellite communication system, however, suffers from the following disadvantages: 1. The initial cost of placing a satellite into its orbit is very high.
2. Owing to the travel of microwave signal from the sending earth station to the satellite and then back to the receiving earth station, there is a 270 msec propagation delay between the sender and receiver of any information. 3. The manner in which a satellite communication system works, any data sent to the satellite for transmission automatically gets broadcast to all receiving stations within the satellite's range. Hence, special security measures are needed to prevent unauthorized tampering of information. 4. Atmospheric disturbances like thunder and lightening affect the K u-band transmission of a satellite communication system. Hence C-band transmission is recommended during bad weather. Optical Fibers Optical fibers are hair-thin threads of glass or plastic that can serve as a data transmission medium as copper wires or coaxial cables. The basic difference is that optical fibers transmit light signals instead of electrical signals. Because light travels much faster than electricity, optical fibers can transmit data at much higher speed than copper wires or coaxial cables, and also with no significant loss of intensity over very long distances. Optical fibers are made of glass, plastic or silica. Plastic fibers are least efficient, but tend to be cheaper and more rugged. Glass or silica fibers are much smaller, and their lower attenuation makes them more suited for very high capacity channels. Physically, a fiber-optic cable consists of three concentric layers - the inner core, a cladding around it, and the outer protective coating. The inner core, which has a diameter of 8 to 200 micrometers, consists of a bunch of optical fibers. The cladding around it is made of plastic or glass and has a refractive index less than that of the core. The characteristics of light propagation depend primarily on the fiber size, its construction, the refractive index profile, and the nature of the light source. The outer protective coating is made up of plastic. The main components of an optical fiber communication system are shown in Figure 17.7. Towards its source side is a converter that converts electrical signals into light waves. The converter uses either a light-emitting diode (LED) or a laser diode to convert electric signals into light signals. These light waves are then transmitted over the optical fiber to the receiver's end. At the receiver's end, another converter is placed that detects the light waves and converts them back to electrical signals. It uses photoelectric diodes for this purpose. These electric signals are then amplified using an amplifier and sent to the receiver. Optical fibers have the following advantages:
1. Large bandwidth. Optical fibers can transmit massive amounts of data at very high speed and long distances. They can have bandwidths 10 to 100 times greater than the best coaxial cables. Their bandwidths are typically in the range of 200 MHz per Km to 3 GHz per Km. A fiber-optic cable consisting of a bunch of 10 optical fibers can accommodate more than 200,000 voice channels. 2. Low loss. As compared to electrical signals, light signals can travel at very high speed with no significant loss of intensity over long distances. In case of use of data transmission medium that use electrical signals for transmitting data, such as copper wires or coaxial cables, electrical signals have to be intercepted and amplified by repeaters at short intervals along the way. Light signals flowing through a fiber-optic cable also must be amplified, but much less so that of electrical signals. The implication is that signal repeaters need not be placed so close together, which reduces the cost of an optical fiber communication system. 3. Immunity to electromagnetic interference. Because they are made of insulators, optical fiber transmissions are not affected by magnetic or electrical interference that can cause errors in other media, especially in twisted-pair wire. Consequently, optical fiber communication systems have the potential for reducing costs associated with complex error checking and correction mechanisms that must be used with media that are affected by electromagnetic interference. 4. Small size and light weight. Fiber optic cables are much smaller in size and lighter in weight than copper wires or coaxial cables. For example, a fiber optic cable with a core and cladding of 0.125 mm diameter and 3.5 mm outer protective coating has the same information carrying capacity as 900 twisted copper wire pairs with an 8 cm outer diameter and weighing 100 times as much. Size and weight are important factors when considering conduits running under overcrowded city streets. In some large cities there simply is no room for additional bulky copper wires. In this case, the use of smaller and lighter fiber optic cables is very effective. 5. Security. Optical fiber offers increased security against unauthorized tampering of information, since it is extremely difficult and expensive to tap optical signals. 6. Safety and electrical insulation. Optical fibers being insulators provide electrical isolation between the source and the destination. Hence they present no electrical spark hazards and can be safely used in those places where it is unsafe to use electrical conductors. 7. Analog and digital signals transmission. Both analog and digital signals can be transmitted by the use of optical fibers. In analog transmission, the light intensity is varied continuously. On the other hand, in digital transmission the light source is turned on or off. Optical fibers, however, suffer from the following disadvantages:
1. Optical fibers being fragile cannot be easily turned at sharp corners. They can turn only those corners that have radius of at least a few inches. This creates problem in the physical laying of the fiber optic cable. 2. Aligning and joining two fiber optic cables is not so simple and easy as for twisted copper wire pairs or coaxial cables. It requires special equipment to do so. Optical fibers are already being widely used for data transmission among computers. Their advantage is making them the transmission medium of choice for many applications involving telephones, televisions and computers. Digital and Analog Data Transmission Data is propagated from one point to another by means of electrical signals, which may be in digital or analog form. As shown in Figure 17.8(a), in an analog signal, the transmitted power varies over a continuous range, for example sound, light and radio waves. The amplitude (v) of analog signals is measured in volts and its frequency (f) in hertz (Hz). The higher is the frequency of the signal, the more number of times it crosses the time axis. However, a digital signal is a sequence of voltage pulses represented in binary form (see Figure 17.8(b)). Computer generated data is digital whereas the telephone lines used for data communication in computer networks are usually meant for carrying analog signals. When digital data is to be sent over an analog facility, the digital signals must be converted to analog form. The technique by which a digital signal is converted to analog form is known as modulation. The reverse process, that is the conversion of analog signal to its digital form at a destination device, is called demodulation. Modulation Techniques Modulation is most often used for superimposing digital data on analog waves. There are three forms modulation - amplitude, frequency and phase modulation. They are discussed below. 1. Amplitude Modulation (AM). Two binary values (0 and 1) of digital data are represented by two different amplitudes of the carrier signal keeping the frequency and phase constant. On voice grade lines, it is used up to 1200 bits per second. However, amplitude modulated signals are sensitive to impulse noises that arise due to electrical sparks near the transmission line. 2. Frequency Modulation (FM). Two binary values of digital data are represented by two differ frequencies while the amplitude and phase are kept constant. Also known as frequency shift key, (FSK), this method is less susceptible to error than amplitude modulation. Another advantage oft method is that it is easier to design devices based on FSK because discrimination between
two frequencies is simpler than detecting phase changes. Thus for medium speed communications (1200 to 2400 bits per second) the FSK scheme is preferred. 3. Phase Modulation (PM). Two binary values or digital data are represented by the shift in phase of the carrier signal. That is, a sine wave with phase = 0° represents a digital 1 and a sine wave with phase = 180° represents a digital 0. This technique is more noise resistant and efficient than both amplitude modulation and frequency modulation. For higher speed transmission of 9600 bits per second, phase modulation is more reliable and is preferred to other methods. Modems The process of modulation and demodulation, that is, the conversion of digital data to analog form and vice-versa, is carried out by a special device called a modem (modulator/demodulator). Hence, when an analog facility is used for data communication between two digital devices (say a terminal and a computer), two modems are required, one near each digital device. As shown in Figure 17.9, the digital signal generated at the terminal is converted to analog form by the modulator of the modem placed near it. The analog signal is transmitted through the telephone line, which is converted to digital form by the demodulator of the modem placed near the computer. This digital data is processed by the computer. The processed digital data is modulated to analog form and returned via the telephone line to the terminal where the analog signals are demodulated to digital form for display on the terminal. Thus, the modem is an essential piece of hardware for any application in which two digital devices (say two computers) want to communicate over an analog transmission channel (say a telephone line). When you want to use a modem with your computer to allow it to communicate with any other computer via the telephone line, the following factors should be considered: 1. Transmission speed 2. Internal versus external 3. Facsimile facility 4. Error correction 5. Data compression 6. Intelligence They are described below. 1. Transmission speed. The higher is the transmission speed of a modem, the better it is because it can communicate faster. The transmission speeds of earlier modems were 300, 1200, or 2400 bps (bits per second). Although modems having transmission speed of 2400 bps can still be purchased, you will not find a new modem that is slower than this.
For achieving transmission speeds higher than 2400 bps, many techniques are used by modem manufacturers. In order to ensure that modems are inter-operable (can communicate with another modem, even if it is made by a different manufacturer), modems must conform to standards called modulation protocols. The standards for higher transmission speeds for modems have been set by the Consultative Committee for International Telephony and Telegraphy (CCITT). The first such standard, called V.32, achieves transmission speeds of 9600 bps (four times the speed of a 2400 bps modem). Another standard, called V.32bis ("bis" means in addition to. Therefore V.32bis modems provide additional functionality to V.32 modems), allows modems to exchange data at a rate of 14400 bps. A more recent standard, called V.34, allows modems to exchange data at a rate of 28800 bps. Note that two modems can communicate with each other only if both follow the same modulation protocol. That is, even if you have a modem that follows the V.34 protocol, it cannot communicate at 28800 bps unless the modem on the other end also follows the same protocol. To take care of this problem, most modems are designed to have fall back feature. This feature allows a fast modem to communicate with any slower modem. That is, when two modems attempt to establish a connection for communication, they try to establish which protocols they share, and then they use the fastest rate that both modems have. That is, if two computers, one with 9600 bps modem and the other one with 14400 bps modem, try to communicate, the data would be transferred between the two computers at 9600 bps. Hence, a fast modem just provides you the option to communicate faster when you can. 2. Internal versus external. Modems are of two kinds, internal and external. An internal modem is an optional add-on circuit board that plugs into one of the computer's expansion slots. It gets its power from the computer's expansion bus. It is manufactured and supplied by the computer manufacturer itself. An external modem, on the other hand, is a separate box that contains the circuitry and logic to modulate data signals. It has its own power supply, on/off switch, and front-panel LCDs to indicate its status. For this reason, external modems are slightly more expensive. An external modem is connected to the computer via a serial port. Both the internal as well as external modems connect to the telephone system with a standard telephone jack. However, each one has its own advantages and disadvantages. Internal modems are slightly faster and do not have external wires and power cords hanging around. However, as they plug into one of the computer's expansion slots, this may be a problem if free slots are scarce. External modems are slightly costlier and have external wires and power cords. They also need one of the built-in serial ports of the computer, which may be a problem if
there are few other devices that also require a serial port for connectivity. However, external modems have the flexibility of being switched off independent of the computer, and if there is any problem with the modem hardware/software, the computer need not be sent for repair and maintenance (only the modem can be sent) 3. Facsimile facility. Some modems, known as FAX modems, are capable of emulating a FAX machine in addition to performing the functions of a regular modem. A computer equipped with a FAX modem can send/receive text and images as a FAX to/from a remote FAX machine or another computer equipped with a FAX modem. Like other kinds of modems, FAX modems can be external or internal. 4. Error correction. Data transmission errors may occur while data moves on telephone lines from one modem to another. To take care of this problem, many of the more sophisticated modems use methods of recovering from data transmission errors. The two standard methods (protocols) used in the design of modems for error correction feature are MNP4 (Microcom Network Protocl Version 4) developed by Microcom, Inc. and V.42 standard developed by CCITT. MNP4 error-correction protocol is suitable for low-speed modems (such as those operating at 2400 bps), whereas V.42 standard is suitable for higher-speed modems (those operating at 9600 bps and 14400 bps). 5. Data Compression. Data compression techniques (discussed in Chapter 19) help in reducing the volume of the data. The reduction in data volume depends mostly on the kind of data that is being compressed and the compression algorithm used, but can be as high as 10:1 ratio. By reducing the volume of the data to be transmitted, modems with built-in data-compression techniques can achieve higher throughput (can transfer more data in the same amount of time) than conventional modems of the same speed. Keeping interoperability in mind, the two standard data compression techniques used in the design of modems are Microcom's MNP Class 5 specification and CCITT's V.42bis specification. MNP5 specification is commonly used for lowspeed modems, whereas V.42bis specification is commonly used for high-speed modems. The effective throughputs of the base data rates for the two standard specifications are given in Figure 17.10. Note that whereas MNP5 can almost double the transmission rate of a modem, V.42bis can almost quadruple it. Of course, for data compression to work effectively, two communicating modems must support a common data-compression standard. Most modems can determine and adjust to the data compression capabilities of another modem automatically. Base Data Rates 2,400 bps 9,600 bps
Throughputs with'Data Compression With MNP5 With V.42bis 4,800 bps 9,600 bps 19,200 bps 38,400 bps
14,400 bps
28,800 bps
57,600 bps
Figure 17.10. Effective throughputs of base data rates with the two standard data compression specifications used in modems. 6. Intelligence. Some of the more expensive modems contain microprocessors, storage chips, and specialized communication chips that allow them to operate and function under a large number of different circumstances. These modems can be programmed to automatically perform dialing, answering, and disconnecting functions. For instance, some modems can automatically dial-up a remote computer (auto-dial), even answer incoming calls from other computers (autoanswer). Similarly, some modems have callback security feature that provides a mechanism of protecting its computer from unauthorized access. These modems include a list of valid phone number and password combinations with routines for automatic passward check and callback. As soon as a remote computer tries to establish a connection with a computer having this type of modem, the modem calls back the remote computer immediately to ensure that the connection is indeed established with the authorized remote user. Modems having such special features are called smart modems or intelligent modems. Analog versus Digital Transmission If data is sent over long distances, the analog signal can become weak and distorted. To avoid this, amplifiers are used. An afnplifier'used in a communication system is similar to the amplifier used in a home stereo system - it strengthens or amplifies a weak signal. Amplifiers are typically placed along communication lines between modems. Today, signals are also sent in digital mode. In this case modems are not needed. When digital mode transmission is used over long distances, a repeater is used to receive and then transmit the signal. Like amplifier, the purpose of & repeater is to strengthen a weak digital signal over long distances. Digital transmission of digital data is preferred to analog transmission of digital data due to the following reasons: 1. Lower cost, 2. No modems required, 3. Higher transmission speeds possible, and 4. Lower error rate (higher reliability). DATA TRANSMISSION SERVICES We saw that there are many different types of data transmission media that may be used for transmission of from one point to another. In practice, transmission of
data from one computer system to another often involves several data transmission media, which together constitute the data transmission path between the sender and receiver computers. For example, the data transmission path between two computers located far apart might be constituted of copper wire, fiber optic cable and communication satellite. Organizations may afford to lay their own communication channels in a smaller geographical area such as within a building or a campus. However, it becomes impractical for organizations to do so in a larger geographical area such as between two cities or two countries. It is also impractical for them to set their own satellites in orbit. Therefore, organizations hire the services of data transmission service providers (popularly known as common carriers) for their data communications needs. VSNL (Videsh Sanchar Nigam Limited), BSNL (Bharat Sanchar Nigam Limited) and MTNL (Mahanagar Telephone Nigam Limited) are a few such common carriers in India, whereas AT&T, MCI, Western Union and GTE are a few such common carriers in the USA. The various types of services offered by the common carriers are: 1. Dial-up line. Dial-up line (also known as switched line) is a service that operates in a manner similar to a telephone call. That is, a user of a computer willing to communicate with a remote computer, first makes a connection request by dialing up the remote computer. A circuit is then established between the two computers via the telephone company's switching system. The modem attached to the user's computer then sends and receives data over the telephone line. Just as in the case of a telephone call, the charge for data transmission service in this case depends on the duration of communication (circuit establishment) and the distance between the two computers. 2. Leased line. Leased line (also known as private line or dedicated line) is a special conditioned telephone line that directly and permanently connects two computers. A leased line can be used for both voice and data transmissions. Hence, if an organization has two offices in two different cities, it is often cheaper for the organization to acquire its own leased line between the two offices which can be used by the organization for all its voice (telephone calls) and data transmissions between the two offices. The charges for a leased line are often based on channel capacity (bps) and distance (air miles). 3. Integrated Services Digital Network (ISDN). The ISDN is a telephonic system that provides digital (not analog) telephone and data services. As it supports digital services (including digitized voice), the ISDN telephone users enjoy noise-free, CD-quality sound. Moreover, with the ISDN, no modem is necessary because it supports digital transmission of all types of data (including voice). This also results in very short call set-up time between two ISDN subscribers.
The main factor that led to the development of the ISDN was the adoption of digital transmission by many public telephone systems. Voice traffic in such networks is encoded using pulse-code modulation at 64 Kbps. Naturally, these networks could also carry data, resulting in integrated digital networks (IDN). Based on the transmission and switching capabilities, the ISDNs are currently of two types - narrowband ISDN and broad-band ISDN. Narrow-band ISDN is based on 64 Kbps bit-streams that are combined into higher-capacity "trunks" using time-division multiplexing. For example, 32 64-Kbps channels can be combined into one 2-Mbps channel. The narrow-band ISDN, however, cannot support the requirements of several types of data services, especially those needed for multimedia applications. For example, the bandwidth required for full-definition digital video is in the 100 Mbps range. To handle this sort of traffic, as well as bursts of data traffic from computers, broad-band ISDN (often referred to as BISDN) was introduced. B-ISDN is based on optical fibers and asynchronous timedivision multiplexing. The advantage of asynchronous time-division multiplexing over conventional time-division multiplexing is that it allows the total bandwidth available to be divided between contending activities in a much more flexible way. ISDN has become so popular that a term to describe the standard analog service has been coined: Plain Old Telephone Service (POTS). 4. Value Added Network (VAN). Some companies specialize in providing value added data transmission service. The value added over and above the standard services of the common carriers may include electronic mail, data encryption/decryption, access to commercial databases, and code conversion for communication between incompatible computers. These companies normally lease dedicated lines of a common carrier, do value addition to enhance the communication facilities, and then sell that enhanced service. This type of service is popularly known as value-added network (VAN) service. COMMUNICATION PROCESSORS In a data communications network, the task of network designers is to select and coordinate the network components so that the necessary data is moved to the right place, at the right time, with a minimum of errors, and at the lowest possible cost. A number of communication processors (typical micro- or mini- computers) are used by network designers to achieve these goals. The functions of some of the commonly used communication processors are briefly described below. Multiplexers
There are many applications in which several terminals are connected to a computer. If each terminal is operating at 300 bits per second over a communication line (channel) that can operate at 9600 bits per second, then we see a very inefficient operation. It has been found that the capacity of a channel exceeds that required for a single signal. A channel is an expensive resource. Hence for its optimal utilization, the channel can be shared in such a way so as to simultaneously transmit multiple signals over it. The method of dividing a physical channel into multiple logical channels, so that a number of independent signals may be simultaneously transmitted on it, is known as multiplexing. The electronic device that performs this task is known as a multiplexer. A multiplexer takes several data communication lines or signals and converts them into one data communication line or signal at the sending location. For example, as shown in Figure 17.11, there may be 4 terminals connected to a multiplexer. The multiplexer takes the signals from the 4 terminals and coverts them into one large signal, which can be transmitted over one communication line. Then, at the receiving location, a multiplexer takes the one large signal and breaks it into the original 4 signals. Without multiplexers, you would have to have 4 separate communication lines. Thus, with multiplexing it is possible for a single transmission medium to concurrently transmit data between several transmitters and receivers. There are two basic methods of multiplexing channels. They are frequency-division multiplexing and time-division multiplexing. 1. Frequency-Division Multiplexing (FDM). The bandwidth or range of a medium exceeds that of any given signal. This fact is utilized for frequency-division multiplexing. In FDM, the available bandwidth of a physical medium is split up (divided) into several smaller, disjoint logical bandwidths. Each of the component bandwidths is used as a separate communication line (channel). Figure 17.12 illustrates the process of FDM. The best example of FDM is the way we receive various stations in a radio. Each radio station is assigned a frequency range within a bandwidth of radio frequencies. Several radio stations may be transmitting speech signals simultaneously over the physical channel, which is "ether" in this case. A radio receiver antenna receives signals transmitted by all the stations. Finally, the tuning dial in the radio is used to isolate the speech signal of the station tuned. In FDM, the signals to be transmitted must be analog signals. Thus, digital signals must be converted to analog form if they are to use FDM. 2. Time-Division Multiplexing (TDM). The bit rate of transmission medium always exceeds the required rate of the digital signal. This fact is utilized for time-
division multiplexing. In TDM, the total time available in the channel is divided between several users and each user of the channel is allotted a time slice (a small time interval) during which he/she may transmit a message. That is, multiple data streams belonging to different users are interleaved into one data stream at the sending end of the channel. At the receiving end, a demultiplexer is used to reassemble individual chunks of messages sent by the users into full messages, Figure 17.13 shows multiplexing of three different signals using TDM. The same idea may be extended for multiplexing several signals. Although TDM may be used to multiplex digital or analog signals, its usage is more appropriate for digital data transmission. It is also very much suitable for communication between computers because such communications occur in short fast bursts. Besides this, TDM is generally more efficient as more subchannels can be derived. It is up to the network designers to allocate time slots to different channels. It is common to have 32 low-speed terminals connected to one highspeed line. Whether or not to use multiplexing usually depends upon economics. The cost of high-speed modems and multiplexers is very high compared to the cost of lowspeed modems. However, if line costs are high due to long distances, then multiplexing is cost effective. One serious disadvantage with multiplexing relates to a transmission line failure. If the line goes out, everything is dead. With individual lines only one terminal is likely to be lost. Concentrators A concentrator performs the same function as a multiplexer, but concentrators have the ability to actually reduce the number of signals. For example, 100 signals from different devices coming into the concentrator could leave as only 70 or 80 signals. The signals are concentrated to form a fewer number of signals. This requires intelligence. This intelligence normally takes the form of microprocessors or even minicomputers. Thus, a concentrator is basically an intelligent multiplexer. The advantage of using a concentrator is that devices of varying speeds and types are connected to the concentrator, which in turn is connected to the host computer by high-speed lines. Concentrators are especially useful where data communication cost is high, such as long-distance international communications. In general, concentrators are more expensive than multiplexers. For some longdistance and international data communication applications, however, the use of concentrators is cost justified. Front-End Processors (FEP)
If information is to be moved from point-to-point in a computer network, then some type of control is necessary. Network processing (also known as communications processing) is the control of information movement between the various locations of the network. This is a substantial task. In the early network days, the communications processing job was handled by the main computer. This began to place heavy additional loads on the CPU. Thus, the ability of the computer to do information processing, i.e., the job for which it was intended, was seriously degraded. As a result, manufactures introduced the frontend processor. This is a separate processor designed specifically to handle the network processing task (see Figure 17.14). A front-end processor is usually installed in the same location as the main computer. Its primary purpose is to offload communications processing task from the main computer, thereby the main computer can be dedicated for applications and data processing jobs. It virtually acts like an intelligent terminal or like a secretary/receptionist to the main computer. ASYNCHRONOUS AND SYNCHRONOUS TRANSMISSION Data transmission on a communication line is normally carried out in two different modes - asynchronous and synchronous. Asynchronous transmission is often referred to as start-stop transmission because of its nature, that is, the sender can send a character at any convenient time and the receiver will accept it. This is the characteristic of many terminals. When a terminal is connected to a computer and an operator manually presses keys on the terminal, the time spent between successive keystrokes would vary. Thus in asynchronous transmission, data is transmitted character by character at irregular intervals. In order to enable the receiver to recognize a character when it arrives, the transmitter "frames" each character. Preceding the character is a start bit and following the character will be one or two (depending upon the system) stop bits. Thus for the 7-bit ASCII code, for each character the transmitter transmits the seven character bits, one parity bit, one start bit, and one or two stop bits for a total of either 10 or 11 bits. The concept of the character framed by start-stop bits is illustrated in Figure 17.15(a). Note that in asynchronous transmission, the time between transmitting any two characters can be zero or of any length. Thus the computer might send a continuous stream of characters to the terminal, or characters keyed in by a terminal operator can be sent to the computer as and when the operator keys in a new character.
Asynchronous transmission is well suited to many keyboard type terminals. The advantage of this method is that it does not require any local storage at the terminal or the computer as transmission takes place character by character. Hence it is cheaper to implement. However, since the transmission line is idle during the time intervals between transmitting characters, the idle time of the transmission line may be a matter of concern if the line cost is high. Even though less efficient than synchronous transmission, it is also used with devices such as printers simply to reduce cost. Synchronous mode of data transmission involves blocking a group of characters in somewhat the same way records are blocked on magnetic tape. Each block is then framed by header and trailer information. The header consists of synchronizing information, which is used by the receiving device to set its clock in synchronism with the sending end clock. The header also contains information to identify the sender and the receiver. Following the header is a block of characters that contains the actual message to be transmitted [see Figure 17.15(b)]. The number of characters in a block may be variable and may consist of hundreds of characters. The message characters in the block are terminated by a trailer. The trailer contains an end of message character followed by a check character to aid detection of any transmission error. Thus with synchronous transmission, entire blocks of characters are framed and transmitted together. Synchronous transmission is well suited to remote communication between a computer and such devices as buffered terminals and printers. Here the logical data block length would be 80 or 132 characters. It is also used for computer-tocomputer communication. The primary advantage of synchronous transmission is its efficiency. It enables the use of much higher data rates than with asynchronous transmission because it eliminates the need for individual start-stop bits on each character. The main disadvantage of synchronous transmission is the need for local buffer storage at the two ends of the line to assemble blocks and also the need for accurately synchronized clocks at both ends. As a result, synchronous equipment usually costs more. SWITCHING TECHNIQUES In its simplest form, data communication takes place between two devices that are directly connected by some form of transmission medium - twisted wires, coaxial cables, microwave and satellite links. Often however, it is impractical or uneconomical for two devices to be directly connected. Instead, communication is achieved by transmitting data from source to destination through a network of intermediate nodes. These nodes provide a switching facility that moves data from node to node until the destination is reached. There are three different methods of
establishing communication links between the sender and receiver in a communication network, namely, circuit switching, message switching and packet switching. They are briefly described below. Circuit Switching It is the simplest method of data communication in which a dedicated physical path is established between the sending and receiving stations through the nodes of the network. This method is used to connect two subscribers for a telephone conversation. Network computers and terminals connected to a telephone network also use this method of establishing communication path among them. The method of circuit switching is illustrated in Figure 17.16. Each rectangle represents a switching node of the communication network. When a message is to be communicated, a physical path is established between the sending station and the receiving station by physically connecting the incoming and outgoing line of each of the intermediate switching nodes that fall on the path. Once a circuit is established between the two stations, it is exclusively used by the two parties and the dedicated physical link between both ends continues to exist until the call is terminated either by the sender or the receiver. As soon as the connection is terminated by one of the two stations, the dedicated resources are deallocated and can now be used by other stations also. Thus circuit switching involves three phases - circuit establishing, data transfer and circuit disconnection. It is used in the Public Switched Telephone Network (PSTN). Advantages 1. The main advantage of circuit switching is that once the circuit is established, data is transmitted with no delay other than the propagation delay, which is negligible (typically of the order of 6 msec per 1000 kms). 2. It is suitable for low-speed communication between a terminal and a computer because the method is simple and requires no special facilities. 3. Since a dedicated continuous transmission path is established, the method is suitable for long continuous transmissions. 4. Since the full capacity of the circuit is available for exclusive use by the connected pair of nodes, the transmission time required to send a message can be known and guaranteed after the circuit has been successfully established. Hence it is the preferred method for transmission of voice and real-time data. Disadvantages
1. Prior to actual data transfer, the time required to establish a physical link between the two stations is of the order of 10 sees or more depending on the distance. For many computer applications, such as point of sale verification, this elapse time is too large and unsuitable. 2. Since the physical link is a dedicated one, the channel capacity is dedicated to one source for the entire duration of a connection. Hence, the network resources are not properly utilized. 3. Since communications between computers occur in short fast bursts with long silent periods in between (during which the communication line is not used), the method proves to be very uneconomical when used for communication between computers that are connected using expensive, high-speed transmission lines. Message Switching A message is a logical unit of information and can be of any length. In this method, if a station wishes to send a message to another station, it first appends the destination address to the message. After this, the message is transmitted from the source to the destination either by store-and-forward method or broadcast method. As shown in Figure 17.17, in the store-and-forward method, the message is transmitted from the source node to an intermediate node. The intermediate node stores the complete message temporarily, inspects it for errors, and transmits the message to the next node based on an available free channel and its routing information. The actual path taken by the message to its destination is dynamic as the path is established as it travels along. When the message reaches a node, the channel on which it came is released for use by another message. In Figure 17.16, if a message is to be transmitted from station A to station B, it may take either path 1-2-3-4 or 1-5-4 depending on the availability of a free output path at that particular moment. As shown in Figure 17.18, in the broadcast method, the message is broadcast over a common medium known as broadcast channel. All the stations check the destination address of each message as they pass by and accept only those addressed to them. The routing delays inherent in store-and-forward method are eliminated in this method. However, this method requires that all stations must be connected to the broadcast channel. Advantages 1. In message switching, no physical connection is required between the source and the destination as required in case of circuit switching.
2. As channels are used only when, messages are transmitted, this method uses the channels very effectively. Disadvantages 1. As the message length is unlimited, each switching node must have sufficient storage to buffer messages. In many cases, this storage space may be underutilized. 2. In message switching, a message is delayed at each node for the time required to receive the message plus a queuing delay waiting for an opportunity to retransmit the message to the next node. 3. The method is too slow for interactive/real-time applications. Hence, it is used when message delays are not critical. Packet Switching This method works in a similar fashion as message switching. However, it overcomes the disadvantages of message switching technique because in this method routing is done on 'packet' basis and not on 'message' basis. A message is split up into 'packets' of a fixed size (of the order of one or few thousand bits). Besides the block of data (part of a message) to be sent, a packet has a header that contains the destination and source addresses, control information, message number, number of current and last packet, synchronization bits, acknowledgement and error checking bytes, etc. Like message switching, the packets may be routed from the sender node to the destination node either by store-and-forward method or by broadcast method. In the store-and-forward method, the packets are stored temporarily at the intermediate nodes for error recovery purposes and are routed to the next node based on an available free channel. The actual path taken by the packet to its destination is dynamic as the path is established as it travels along. Thus, it is possible that different packets of the same message can be routed across different paths leading to the same destination depending upon line availability and congestion. Packet switching is used in the X.25 public packet network and the Internet. Advantages 1. Unlike messages, packets are of small and fixed size. Hence the storage requirement for buffering the packets at intermediate nodes is minimal. 2. Since the routing is done on packet basis, the intermediate nodes do not have to wait for the entire message, hence the transmission is very fast. 3. The method is fast enough for interactive/real-time applications.
4. It is most suitable for "bursty" computer-to-computer communication and is widely used in wide area networks. Disadvantages 1. Due to the need to buffer each packet at every intermediate node and to reassemble the packets at the destination node, the overhead incurred per packet is large. 2. There is no guarantee of how long it takes a message to go from its source node to its destination node because the time taken for each packet depends on the route chosen for that packet, along the with volume of data being transferred along that route. ROUTING TECHNIQUES When multiple paths exist between the source and destination nodes of a packet, any one of the paths may be used to transfer the packet. For example, in the network configuration of Figure 17.17, there are two paths between nodes A and B: 1-5-4 and 1-2-3-4 and any one of the two may be used to transmit a packet from node A to B. The selection of the actual path to be used for transmitting a packet is determined by the routing technique used. An efficient routing technique is crucial to the overall performance of the network. A good routing algorithm should make routing decisions as fast as possible and should be easily implementable in hardware. Furthermore, the decision process usually should not require global state information of the network because such informal gathering is a difficult task and creates additional traffic in the network. Routing algorithms are usually classified based on the following two attributes: 1. Place where routing decisions are made, and 2. Time constant of the information upon which the routing decisions are based. Note that routing techniques are not needed in LANs (described later in this chapter) because the sender of a message simply puts the message on the communication channel and the receiver takes it off from the channel. There is no need to decide the path to be used for transmitting the message from the sender to the receiver. Place Where Routing Decisions are Made Based on this attribute, routing algorithms may be classified into the following three types [Reprinted, with permission, from the book titled Distributed Operating Systems: Concepts and Design by Pradeep K. Sinhi ©1997 IEEE]: 1. Source routing. In this method, the source node selects the entire path before sending the packet. That is, all intermediate nodes via which the packet will be
transferred to its destination are decided at the source node of the packet, and this routing information is included along with the packet. The method requires that the source node must have fairly comprehensive information about the network environment. However, the routing decision process is efficient because the intermediate nodes need not make any routing decision. A drawback of the method is that the path cannot be changed after the packet has left the source node, rendering the method susceptible to component failures. 2. Hop-by-hop routing. In this method, each node along the path decides only the next node for the path. That is, each node maintains information about the status of all its outgoing channels and the adjacent nodes and then selects a suitable adjacent node for the packet and transmits it to that node. The routing decisions are typically based on the channel availability and the readiness of the adjacent nodes to receive and relay the packet. The method requires that each node must maintain a routing table of some sort. However, as compared to the source routing method, this method makes more efficient use of network bandwidth and provides resilience to failures because alternative paths can be used for packet transmissions. 3. Hybrid routing. This method combines the first two methods in the sense that the source node specifies only certain major intermediate nodes of the complete path, and the subpaths between any two of the specified nodes are decided by the method of hop-by-hop routing. Static and Dynamic Routing Depending on when the information used for making routing decisions is specified and how frequently it is modified, routing algorithms are classified into the following two types [Reprinted, with permission, from the book titled Distributed Operating Systems: Concepts and Design by Pradeep K. Sinha. © 1997 IEEE]: 1. Static routing. In this method, routing tables (stored at the nodes) are set once and do not change for very long periods of time. They are changed only when the network undergoes major modifications. Static routing is also known as fixed or deterministic routing. Static routing is simple and easy to implement. However, it makes poor use of network bandwidth and causes blocking of a packet even when alternative paths are available for its transmission. Hence, static routing schemes are susceptible to component failures. 2. Dynamic routing. In this method, routing tables are updated relatively frequently, reflecting shorter-term changes in the network environment. Dynamic routing strategy is also known as adaptive routing because it has a tendency to adapt to the dynamically changing state of the network, such as the presence of faulty or congested channels. Dynamic routing schemes can use alternative paths for packet transmissions, making more efficient use of network bandwidth and providing resilience to failures. In dynamic routing, however, packets of a message may arrive out of order at the destination node. This problem can be
solved by appending a sequence number to each packet and properly reassembling the packets at the destination node. NETWORK TOPOLOGIES A network is a logical extension of a data communication system. In a computer network, two or more computers (often referred to as nodes) are linked together with carriers and data communication devices for the purpose of communicating data and sharing resources. The term network topology refers to the wav in which the nodes of a network are linked together. It determines the data paths that may be used between any pair of nodes in the network. Although the number of possible network topologies is seemingly limitless, the four major ones are the star network, the ring network, the completely connected network, and the multiaccess bus network. The choice of network topology for installing a computer network depends upon a combination of factors such as: 1. The desired performance of the system. 2. The desired reliability of the entire system. 3. Size (number of nodes and their geographical distribution) of the system. 4. Expandability of the system. 5. Cost of the components and services required to implement the network. 6. Availability of communication lines. 7. Delays involved in routing information from one node to another.
Star Network Figure 17.19 shows the star arrangement of a computer network. In this configuration, multiple computers are connected to a host computer. That is, the computers in the network are not linked directly to each other and can communicate only via the host computer. The routing function is performed by the host computer, which centrally controls communication between any two other computers by establishing a logical path between them. . Advantages 1. Star topology has minimal line cost because only n-1 lines are required for connecting n nodes. 2. Transmission delays between two nodes do not increase by adding new nodes to the network because any two nodes may be connected via two links only.
3. If any of the local computers fails, the remaining portion of the network is unaffected. Disadvantage 1. The system crucially depends on the central node. If the host computer fails, the entire network fails. Ring Network Figure 17.20 shows the circular or ring arrangement of a computer network. In this configuration, each computer in the network has communicating subordinates, but within the ring there is no master computer for controlling other computers. A node receives data from one of its two adjacent nodes. The only decision a node has to take is whether the data is for its own use or not. If it is addressed to it, it utilizes it. Otherwise, it merely passes it on to the next node. Advantages 1. The ring network works well where there is no central-site computer system. 2. It is more reliable than a star network because communication is not dependent on a single host computer. If a link between any two computers breaks down, or if one of the computers breaks down, alternate routing is possible. Disadvantages 1. In a ring network, communication delay is directly proportional to the number of nodes in the network. Hence addition of new nodes in the network increases the communication delay. 2. The ring network requires more complicated control software than star network. Completely Connected Network As shown in Figure 17.21, a completely connected network has a separate physical link for connecting each node to any other node. Thus, each computer of such a network has a direct dedicated link, called a point-to-point link, with all other computers in the network. The control is distributed with each computer deciding its communication priorities. Advantages 1. This type of network is very reliable, as any link breakdown will affect only communication between the connected computers.
2. 3.
Each node of the network need not have individual routing capability. Communication is very fast between any two nodes.
Disadvantage 1. It is the most expensive network from the point of view of link cost. If there are n nodes in the network, then (n-l)/2 links are required. Thus, the cost of linking the system grows with the square of the number of nodes. Multi-access Bus Network Figure 17.22 shows a multi-access bus network. In this type of network, a single transmission medium is shared by all nodes. That is, all the computers are attached to the same communication line (channel). When a particular computer wants to send a message to another computer, it appends the destination address to the message and checks whether the communication line is free. As soon as the line becomes free, it broadcasts (place) the message on the line. As the message travels on the line, each computer checks whether it is addressed to it. The message is picked up by the addressee computer, which sends an acknowledgement to the source computer and frees the line. This type of network is also known as 'multipoint' or 'multidrop' or 'broadcasting' network. It is appropriate for use in a local area network where a high-speed communication channel is used and computers are confined to a small area. It is also appropriate when satellite communication is used as one satellite channel may be shared by many computers at a number of geographical locations.
Advantages 1. The main advantage of a multi-access bus network is the reduction in physical lines. 2. The failure of a computer in the network does not affect the network functioning for other computers. 3. Addition of new computers to the network is easy. Disadvantages 1. All computers in the network must have good communication and decisionmaking capability. 2. If the communication line fails, the entire system breaks down. Hybrid Network
Different network configurations have their own advantages and limitations. Hence in reality, a pure star or ring or completely connected network is rarely used. Instead, an organization will use some sort of hybrid network, which is a combination of two or more different network topologies. The exact configuration of the network depends on the needs and the overall organizational structure of the company involved. In some cases, the hybrid network may have components of star, ring, and completely connected networks. A typical hybrid network of this type is shown in Figure 17.23. NETWORK TYPES (LAN, WAN AND MAN) Networks are broadly classified into two types: local area networks (LANs) and wide-area networks (WANs). The WANs are also referred to as long-haul networks. The key characteristics1 that are often used to differentiate between these two types of networks are as follows [Reprinted, with permission, from the book titled Distributed Operating Systems: Concepts and Design by Pradeep K. Sinha. © 1997 IEEE]: 1. Geographic distribution. The main difference between the two types of networks is the way in which they are geographically distributed. A LAN is restricted to a limited geographic coverage of a few kilometers, but a WAN spans greater distances and may extend over several thousand kilometers. Therefore, LANs typically provide communication facilities within a building or a campus, whereas WANs may operate nationwide or even worldwide. 2. Data rate. Data transmission rates are usually much higher in LANs than in WANs. Transmission rates in LANs usually range from 0.2 Megabit per second (Mbps) to 1 Gigabit per second (Gbps). On the other hand, transmission rates in WANs usually range from 1200 bits per second to slightly over 1 Mbps. 3. Error rate. Local area networks generally experience fewer data transmission errors than WANs do. Typically, bit error rates are in the range of 10-5 to 10-7 with WANs. 4. Communication link. The most common communication links used in LANs are twisted pair, coaxial cable, and fiber optics. On the other hand, since the sites in a WAN are physically distributed over a large geographic area, the communication links used are by default relatively slow and unreliable. Typical communication links used in WANs are telephone lines, and microwave links satellite channels. 5. Ownership. A LAN is typically owned by a single organization because of its limited geographic coverage. A WAN, however, is usually formed by interconnecting multiple LANs each of which may belong to a different organization. Therefore, administrative and maintenance complexities and costs for LANs are usually much lower than for WANs. 6. Communication cost. The overall communication cost of a LAN is usually much lower than that of a WAN. The main reasons for this are lower error rates, simple (or absence of) routing algorithms, and lower administrative and
maintenance costs. Moreover, the cost to transmit data in a LAN is negligible since the transmission medium is usually owned by the user organization. However, with a WAN, this cost may be very high because the transmission media used are leased lines or public communication systems, such as telephone lines, microwave links, and satellite channels. Networks that share some of the characteristics of both LANs and WANs are sometimes referred to as metropolitan area networks (MANs). The MANs usually cover a wider geographic area (up to about 50 km in diameter) than LANs and frequently operate at speeds very close to LAN speeds. The main objective of MANs is to interconnect LANs located in an entire city or metropolitan area. Communication links commonly used for MANs are coaxial cable and microwave links. Ethernet is the most widely used LAN technology because it is relatively fast and economical. It was introduced by DEC (Digital Equipment Corporation), Intel, and Xerox in 1980, and subsequently, a slightly modified version of it was adopted by the IEEE as a standard LAN technology. The network topology used for Ethernet is multi-access bus topology. The communication medium used is low-loss coaxial cable having a data transfer rate of 10 Mbps. A message is transmitted from one node to another by breaking it up into packets (called frames in Ethernet) and then by broadcasting the packets to the bus. An address designator is associated with each packet. As a packet travels on the bus, each node checks whether the packet is addressed to it and the addressee node picks up the message. The ARPANET of the Advanced Research Projects Agency of the U.S. Department of Defense was the first WAN. It connected about 40 universities and research institutions throughout the United States and Europe with about 50 computers ranging in size from mini-computers to supercomputers. ERNET (Education and Research NETwork) is a WAN, which links several education and research institutions in India. NICNET (National Informatics Centre's NETwork) is a WAN, which links several offices of NIC (National Informatics Centre) of India up to district headquarters. Indian Railways also has its own WAN which links its booking offices across the nation facilitating the passengers to book tickets on most major trains from anywhere in India. Similarly, international airlines worldwide also link their computers to form a WAN that facilitates booking of flight tickets on international flights. SWIFT is a WAN that interconnects the computers of International banks to reconcile their accounts and for electronic funds transfer. The Internet (described in the next chapter) is a WAN that encompasses many LANs and computers across the world. COMMUNICATION PROTOCOLS A protocol is a set of formal operating rules, procedures, or conventions that govern a given process. A communication or network protocol, therefore,
describes the rules that govern the transmission of data over communication networks. These rules provide a method for orderly and efficient exchange of data between the sender and the receiver and for the proper interpretation of controls and data transmitted as raw bits and bytes. These rules are embedded in the data communication software. Roles of a Communication Protocol In any computer network, a communication protocol normally performs the following functions for the efficient and error-free transmission of data. It has a separate set of rules (implemented in software) for performing each of these functions. 1. Data sequencing. It refers to breaking a long message into smaller packets of fixed size. Data sequencing rules define the method of numbering (or sequencing) packets to detect loss or duplication of packets, and to correctly identify packets that belong to the same message. 2. Data routing. Routing algorithms are designed to find the most efficient paths between a source and a destination. They can handle varying degree of traffic on the present network configuration with optimal time utilization. 3. Data formatting. Data formatting rules define which group of bits or characters within a packet constitutes data, control, addressing, or other information. 4. Flow control. A communication protocol also prevents a fast sender from overwhelming a slow receiver. It ensures resource sharing and protection against traffic congestion by regulating the flow of data on the communication lines. 5. Error control. These rules are designed to detect errors in messages and to ensure transmission of correct messages. The most common method for correcting errors is to retransmit the erroneous message block. This method requires coordination between the sender and the receiver nodes so that the block having error is discarded by the receiver node and is retransmitted by the sender node. 6. Precedence and order of transmission. These rules condition all nodes about when to transmit their data and when to receive data from other nodes. It is ensured that all nodes get a chance to use the communication lines and other resources of the network depending upon the priorities assigned to them. 7. Connection establishment and termination. These rules define how connections are established, maintained and terminated when two nodes of a network want to communicate with each other. 8. Data security. Providing data security and privacy is also built into most communication software packages. It prevents access of data by unauthorized users. 9. Log information. Several data communication software are also designed to develop log information, which consists of all jobs and data communications
tasks that have taken place. Such information is normally used for charging the various users of the network based on their usage of the network resources. Concept of Layered Protocols in Network Design Most modern computer networks are designed in a modular or structured fashion for easy and efficient handling of the system. They are normally split up into a series of modules and are logically composed of a succession of layers or levels. Each layer offers certain services to the higher layers, shielding those layers from the details of how the offered services are actually implemented. Each layer has its own set of protocols. A particular layer of one machine communicates only with the corresponding layer of another machine using the protocols of this layer. The main reasons for using the concept of layered protocols in network design are as follows: 1. The protocols of a network are fairly complex. Designing them in layers makes their implementation more manageable. 2. Layering of protocols provides well-defined interfaces between the layers, so that a change in one layer does not affect an adjacent layer. That is, the various functionalities can be partitioned and implemented independently so that each one can be changed as technology improves without the other ones being affected. For example, a change to a routing algorithm in a network control program should not affect the functions of message sequencing, which is located in another layer of the network architecture. 3. Layering of protocols also allows interaction between functionally paired layers in different locations. This concept aids in permitting the distribution of functions to remote nodes. The terms protocol suite, protocol family, or protocol stack are used to refer to the collection of protocols (of layers) of a particular network system. Network Interface Cards Network interface card, often referred to as NIC or network card, is a hardware device that allows a computer to be connected to a network, both functionally and physically. The NIC is a printed circuit board that is installed on to one of the expansion slots of the computer and provides a port on the back of the computer to which the network cable is attached. Thus, the NIC is one of the several add-on cards (expansion boards) that the computer may have. As the NIC is connected directly to the computer's I/O bus, the design of a NIC is specific to the computer's I/O bus hardware, the computer's operating system, and the network's communication protocol. The network's communication protocol is embedded in the NIC's ROM. Thus, there are different NICs for different networks
even for the same computer. For example, if a computer is to be connected to an Ethernet LAN, it must be equipped with an Ethernet network card, and if it is to be connected to an ATM network, it must be equipped with an ATM network card. In addition to the communication software embedded in the NIC's ROM, the computer also requires network software that tells the computer how to use the NIC. Both the network software and the NIC have to adhere to the communication protocol of the network to which the computer is to be connected. The OSI Model The initial computer networks had their own set of standards and conventions that were quite hardware oriented. Each manufacturer used to develop their own communication protocols for their networks. For example, IBM launched SNA (Systems Network Architecture) in 1974. Similarly, DEC (Digital Equipment Corporation) launched its network in 1980 for use on the DEC range of computers. The architecture of DECNET (Digital Equipment Corporation Network) is known as DNA (Digital Network Architecture). Like SNA, it is a private network architecture oriented only to DEC hardware. Other telecommunication protocols are BNA (Burroughs Network Architecture) by Burroughs Incorporation, DSN (Distributed Systems Network) by Hewlett Packard, PRIMENET (Prime Computers Network) by Prime Incorporation, etc. Thus the data communication protocols of one network were not compatible with any other network. Moreover, standards of the same network architecture also kept changing from time to time. For example, earlier versions of SNA were not compatible with its subsequent versions. Such incompatibilities started creating bottleneck in the efficient and proper utilization of network resources. This problem was recognized by the International Standards Organization (ISO), which established a subcommittee to develop an international standard on network architectures. The result was the Open System Interconnection (OSI) model, which is a framework for defining standards for linking heterogeneous computers in a packet switched network. Thus the standardized OSI protocols made it possible for any two heterogeneous computer systems, located anywhere in the world, to easily communicate with each other. The OSI model is also designed in a highly structured way. Its architecture is shown in Figure 17.24. It is a seven-layer architecture in which a separate set of protocols is defined for each layer. Thus each layer has an independent function and deals with one or more specific aspects of the communication. The roles of the seven layers are briefly described below. This description material has been reprinted, with permission, from the book titled Distributed Operating Systems: Concepts and Design by Pradeep K. Sinha. [© 1997 IEEE].
In actual implementation, of the seven layers, the first three layers are likely to be in hardware, the next two layers in the operating system, the presentation layer in library subroutines in the user's address space, and the application layer in the user's address space, and the application layer in the user’s program. Physical Layer The physical layer is responsible for transmitting raw bit streams between two nodes. That is, it may convert the sequence of binary digits into electric signals, light signals, or electromagnetic signals depending on whether t two nodes are on a cable circuit, fiber-optic circuit, or microwave/radio circuit, respectively. Electrical details such as how many volts to use for 0 and 1, how many bits can be sent per second, and whether transmission can take place only in one direction or in both directions simultaneously are also decided by the physical layer take place only in one direction or in both directions simultaneously are also decided by the physical layer protocols. In addition, the physical layer protocols also deal with the mechanical details such as the size and shape of the connecting plugs, the number of pins in the plugs, and the function of each pin. In short, the physical layer protocols deal with the mechanical, electrical, procedural, and functional characteristics of transmission of raw bit streams between two sites. RS232-C is a popular physical layer standard for serial communication lines. Data-Link Layer The physical layer simply transmits the data from the sender's node to the receiver's node as raw bits. It is the responsibility of the data-link layer to detect and correct any errors in the transmitted data. Since the physical layer is only concerned with a raw bit stream, the data-link layer partitions it into frames so that error detection and correction can be performed independently for each frame. The data-link layer also performs flow control of frames between two sites to ensure that a sender does not overwhelm a receiver by sending frames at a rate faster than the receiver can process. Therefore, the error control and flow control mechanisms of a network form the data-link layer protocols in the OSI model. Notice that the data-link layer and physical layer protocols establish an error-free communication of raw bits between two nodes. Network Layer The network layer is responsible for setting up a logical path between two nodes for communication to take place. It encapsulates frames into packets that can be transmitted from one node to another using a high-level addressing and routing scheme. That is, routing is the primary job of the network layer and the routing algorithm forms the main part of the network layer protocols of the network.
Two popular network layer protocols are the X.25 Protocol and the Internet Protocol (called IP). The X.25 is a connection-oriented protocol that is based on the concept of establishing a virtual circuit between the sender and receiver before the actual communication starts between them. In this protocol, a request for connection is first sent to the destination, which can either be accepted or rejected. If the connection is accepted, the requesting party is given a connection identifier to use in subsequent requests. During the connection establishment phase, a route between the two parties is also decided that is used for the transmission of subsequent traffic. On the other hand, IP is a connectionless protocol in which no connection is established between the sender and receiver before sending a message. Therefore, each packet of the message is transmitted independently and may take a different route. IP is part of the DoD (U.S. Department of Defense) protocol suite. Notice that the functions performed at the network layer are primarily required in WANs. In a single LAN, the network layer is largely redundant because packets can be transmitted directly from any node on the network to any other node. Therefore the network layer, if present, has little work to do. Transport Layer The job of the transport layer is to provide node-to-node communication and to hide all the details of the communication subnet from the session layer by providing a network-independent transport service. Using this service, all the details of the communication subnet are sealed and one subnet can be replaced with another without disturbing the layers above the transport layer. In particular, the transport layer accepts message of arbitrary length from the session layer, segments them into packets, submits them to the network layer for transmission, and finally reassembles the packets at the destination. Some packets may be lost on the way from the sender to the receiver, and depending on the routing algorithms used in the network layer, packets may arrive at the destination in a sequence that is different from the order in which they are sent. The transport layer protocols include mechanisms for handling lost and out-of-sequence packets. For this, the transport layer records a sequence number in each packet and uses the sequence numbers for detecting lost packets and for ensuring that messages are reconstructed in the correct sequence. The two most popular transport layer protocols are the Transport Control Protocol (TCP) and the User Datagram Protocol (UDP). TCP is a connection-oriented transport protocol. It uses end-to-end mechanisms to ensure reliable, ordered delivery of data over a logical connection. These goals are basically achieved by
using packet sequence numbers and positive acknowledgements with timeout and retransmission. The UDP is a connectionless transport protocol. It is an unreliable protocol because, when it is used, message packets can be lost, duplicated, or arrive out of order. Therefore, only those applications that do not need reliable communication should use UDP. Session Layer The purpose of the session layer is to provide the means by which presentation entities can organize and synchronize their dialog and manage their data exchange. It allows the two parties to authenticate each other before establishing a dialog session between them. It also specifies dialog type - one way, two way alternate, or two way simultaneous - and initiates a dialog session if the message is a connection request message. The other services of the session layer include quarantine service, dialog control, and priority management. The quarantine service buffers a group of messages on the receiving side until the session layer on the sending side explicitly releases them. This is useful in database applications where a transaction (consisting of a group of messages) needs to be an atomic unit. The dialog control is useful for dialog sessions in which the user primitives used for sending and receiving messages are of the nonblocking type. In this case, the user may have multiple requests outstanding on the same session, and replies may come back in an order different from that in which the requests were sent. The dialog control reorders replies according to the order of requests. The priority management service is useful for giving priority to important and time-bound messages over normal, less-important messages. The session layer is not required for connectionless communication. Presentation Layer The purpose of this layer is to represent message information to communicating application layer entities in a way that preserves meaning while resolving syntax differences. For this, the presentation layer may perform one or more of the following types of transformations on message data: 1. A message usually contains structured information that may include any of the data types used in programming languages - integers, characters, arrays, records, and so on, including user-defined data types. Translation is therefore required where language systems or application programs in the source and destination computers use different representations for these data types. 2. Data format conversions are also needed to transfer data between computers when the hardware of the sending and receiving computers uses different data
representations. In this case, the presentation layer software in the sending computer transforms message data from the formats used in its own computer to a set of standard network representations called eXternal Data Representation {XDR) before transmission. The presentation layer software in the receiving computer transforms the message data from the network representations to the formats used in its own computer. 3. For applications dealing with confidential or secret data, the presentation layer software in the sending computer encrypts message data, before passing it to the session layer. On the receiver side, the encrypted message data is decrypted by the presentation layer before being passed on to the application layer. 4. In a similar manner, when message data is large in volume (such as multimedia data) or with networks that are slow or heavily loaded, message data may be compressed and decompressed by the presentation layer software in the sending and receiving computers, respectively. Application Layer The application layer provides services that directly support the end users of the network. Obviously, the functionality implemented at this layer of the architecture is application-specific. Since each application has different communication needs, no fixed or standard set of application layer protocols can meet the needs of all applications. Therefore, the application layer is basically a collection of miscellaneous protocols for various commonly used applications such as electronic mail, file transfer, remote login, remote job entry, and schemas for distributed databases. Some popular application layer protocols are X.400 (Electronic Mail Protocol), X.500 (Directory Server Protocol), FTP (File Transfer Protocol), and rlogin (Remote Login Protocol). Example of Message Transfer in the OSI Model The example below has been reprinted, with permission, from the book titled Distributed Operating Systems: Concepts and Design by Pradeep K. Sinha. © 1997 IEEE. To illustrate the functions of the various layers of the OSI model, let us consider a simple example of message transmission. With reference to Figure 17.25, let us assume that a process at the sending node wants to send a message M to a process at the receiving node. The sending node's process builds the message M and passes it to the application layer (layer 7) on its machine. The application layer software adds a header (H7) to M and passes the resulting message to the presentation layer (6) via the interface between layers 7 and 6. The presentation layer software performs text compression, code conversion, security encryption, and so on, on the received message, and after adding a header (H6) to it, it passes the resulting
message on to the session layer (5). Depending on the type of dialog, the session layer software establishes a dialog between the sender and the receiver processes. It also regulates the direction of message flow. A header (//$) is added to the message at this layer, and the resulting message is passed on to the transport layer (4). The transport layer software now splits the message into smaller units (Mi and M2) called packets and adds a header (H4) to each packet. These headers contain the sequence numbers of the message packets. The packets are then passed on to the network layer (3). The network layer software makes routing decisions for the received packets and sets up a logical path between the sending and receiving nodes for transmission of the packets. It then adds a header (Hi) to each packet and passes them on to the data-link layer (2). The data-Jink layer software adds a header (H2) and a trailer (T2) to each of these packets. The trailers contain the checksum of the data in the corresponding packets. The resulting message units are called frames, which are passed on to the physical layer (1). The physical layer software simply transmits the raw bits from the sender's machine to the receiver's machine using the physical connection between the two machines. On the receiver's machine, the message data traverses up from the physical layer to the application layer. As the message data traverses to higher-level layers, each layer performs the functions assigned to it and strips off the headers or trailers added by its peer layer at the sending node. For example, the data-link layer at the receiving machine performs error detection by recalculating the checksum for each frame and comparing it with the checksum in the trailer of the frame. It strips off the header (Hi) and the trailer (7y from the frames before passing them on to the network layer. The application layer of the receiver's machine finally passes on the message in its original form to the communicating process on the receiver's node. Notice that the software of a particular layer on the sending machine conceptually communicates only with its peer layer on the receiving machine, although physically it communicates only with the adjacent layers on the sending machine. This abstraction is crucial to network design. INTERNETWORKING TOOLS Interconnecting of two or more networks to form a single network is called internetworking, and the resulting network is called an internetwork. Therefore, a WAN of multiple LANs is an internetwork. Internetworks are often heterogeneous networks composed of several network segments that may differ in topology and protocol. For example, an internetwork may have multiple LANs, some of which may have multi-access bus topology while others may have ring topology; some of these LANs may be using Ethernet technology which others may be using Token Ring technology; and some segments of the network may be using the IP suite while others may be using
IBM's SNA (System Network Architecture) protocol suite. Internetworking allows these, relatively unrelated networks to evolve into a single working system. That is, the goal of internetworking is to hide the details of different physical networks, so that the resulting internetwork functions as a single. Interconnecting two networks having the same topology and protocol is simple because the two networks can easily communicate with each other. However, interconnecting two dissimilar networks that have different topologies and protocols requires an internetworking scheme that provides some common point of reference for the two networks to communicate with each other. That point of reference might be a high-level protocol common to the two networks, a device that allows interconnection of different topologies with different physical and electrical characteristics, or a protocol that allows operating environment differences to be ignored. The most commonly used approach is to make use of common high-level protocols for moving data between common layers on a communications model such as the OSI or the IP suites. Internetworking tools, such as bridges, routers, brouters, and gateways, make extensive use of this approach. As described below, each of these tools has strengths, weaknesses, and specific applications in internetworking. These tools are "blackbox" internetworking technologies that enable interconnection of similar or dissimilar networks to form a single network system. Bridges Bridges operate at the bottom two layers of the OSI model (physical and data link). Therefore, they are used to connect networks that use the same communication protocols above the data-link layer but may or may not use the same protocols at the physical and data-link layers. For example, bridges may be used to connect two networks, one of which uses fiber-optic communication medium and the other uses coaxial cable; or one of which uses Ethernet technology and the other uses Token Ring technology. But both networks must use the same high-level protocols (e.g., TCP/IP or XNS) to communicate. The similarity of higher-level protocols implies that bridges do not modify either the format or the contents of the frames when they transfer them from one network segment to another (they simply copy the frames). Hence bridges feature highlevel protocol transparency. They can transfer data between two network segments over a third segment in the middle that cannot understand the data passing through it. As far as the bridge is concerned, the intermediate segment exists for routing purposes only. Bridges are intelligent devices in the sense that they use a process of learning and filtering in data forwarding to keep network traffic within the segment of the network to which it belongs. Therefore, bridges are also useful in network
partitioning. When the performance of a network segment degrades due to excessive network traffic, it can be broken into two network segments with a bridge interconnecting the two segments. Routers Routers operate at the network layer of the OSI model. Therefore, routers do not care what topologies or access-level protocols the interconnected network segments use. Since routers use the bottom three layers of the OSI model, they are usually used to interconnect those networks that use the same high-level protocols above the network layer. Note that the protocols of data-link and physical layers are transparent to routers. Therefore, if two network segments use different protocols at these two layers, a bridge must be used to connect them. Unlike bridges, routers do not view an internetwork from end to end. That is, bridges know the ultimate destination of a data, but routers only know which is the next router for the data being transferred across the network. However, routers are smarter than bridges in the sense that they not only copy a data from one network segment to another, but they also choose the best route for the data by using information in a routing table to make this decision. That is, managing traffic congestion is a big plus of routers; they employ a flow control mechanism to direct traffic on to alternative, less congested paths. Routers are commonly used to interconnect those network segments of large internetworks that use the same communication protocol. They are particularly useful in controlling traffic flow by making intelligent routing decisions. An internetwork often uses both bridges and routers to handle both routing and multiprotocol issues. This requirement has resulted in the design of devices called brouters, which are a kind of hybrid of bridges am routers. They provide many of the advantages of both bridges and routers. They are complex, expensive, and difficult to install, but for very complex heterogeneous internetworks in which the network segments use the same high-level communication protocols, they often provide the best internetworking solution. Gateways Gateways operate at the top three layers of the OSI model (session, presentation, and application). They are the most sophisticated internetworking tools and are used for interconnecting dissimilar networks that use different communication protocols. That is, gateways are used to interconnect networks that are built on totally different communications architectures. For instance, a gateway may be
used to interconnect two networks, one of which uses the IP suite and the other uses the SNA protocol suite. Since networks interconnected by a gateway use dissimilar protocols, protocol conversion is the major job performed by gateways. Additionally, gateways sometimes also perform routing functions. WIRELESSNETWORKS One of the most challenging and interesting recent trends in the computer and communications industries is the integration of wireless communications and computing. The resulting network is referred to as a wireless computing systems. A wireless computing system enhances the functionality of computing equipment by freeing communication from the location constraints of the wired computing systems. Types of Wireless Computing Systems Wireless computing systems can be broadly classified into the following two categories: 1. Fixed wireless systems. These wireless computing systems support little or no mobility of the computing equipment associated with the wireless network. For example, a LAN can be set up using wireless technology to get rid of the hassles of laying cables. The LAN will work as a conventional wired LAN except for the difference that it does not need any cabling to be carried out. 2. Mobile wireless systems. These wireless computing systems support mobility of the computing equipment, which the users use to access the resources associated with the wireless network. In turn, these systems support mobility of users and allow the mobile users to access information-from anywhere and at anytime. The resulting computing environment, which is often called mobile computing or nomadic computing, no. longer requires a user to maintain a fixed position in the network and enables almost unrestricted user mobility. Typical computing equipment used in mobile wireless systems includes smart phones, personal digital assistants (PDAs), and pagers with Internet access. Note that mobile users do not necessarily need to use wireless interfaces and wireless interfaces do not necessarily support mobility. A mobile user can simply connect to fixed networks using wired interfaces as he/she moves. Likewise, a fixed-location user might use a wireless interface (via a LAN) while sitting in an office. Therefore, mobile and wireless computing systems are not the same even though there is considerable overlap.
Issues in Wireless Computing Systems Wireless computing systems communicate by modulating radio waves or pulsing infrared light. Wireless communication is linked to the wired network infrastructure by stationary transceivers. The area covered by an individual transceiver's signal is known as a cell. Cell sizes vary widely. For instance, an infrared transceiver can cover a small meeting room, a cellular phone transceiver has a range of a few miles, and a satellite beam can cover an area more than 400 miles in diameter. The successful use of wireless computing systems faces several challenges. The important ones among them are as follows: 1. Lower bandwidths. Wireless communication is characterized by lower bandwidths. That is, the channel capacity typically available in wireless systems is much lower than what is available in wired networks due to the limited spectrum available, power restrictions, and noise levels. 2. Variable bandwidths. In case of mobile wireless systems, the limited bandwidth that is available is also variable depending on location. Channel reuse must, therefore, be an integral part of the system in order to provide service to as many potential subscribers as possible. 3. Higher error rates. Noise and interference have more impact on wireless computing systems than for wired computing systems. This, in turn, increases communication latency resulting from retransmissions, retransmission time-out delays, and error-correction protocol processing. 4. Increased security concerns. Security is a greater concern in wireless systems than in wired-systems since information may be traveling in free space (with the exception of infrared LANs). Moreover, a stronger level of authentication is needed in a mobile environment where network nodes can potentially join or leave a network at will. 5. Dynamically changing network. In case of a mobile wireless system, the topology of the network changes rapidly due to the movement and resulting connections and disconnections of the mobile nodes. Moreover, the number of mobile nodes in a network cell varies dynamically, and large concentrations of mobile users, say, at conventions and public events, may overload network capacity. 6. Lost or degraded connections. In case of a mobile wireless system, mobility can cause wireless connections to be lost or degraded. Users may travel beyond the coverage of network transceivers or enter areas of high interference. 7. Support for routing and location management functions. Mobile wireless systems also need to provide support for routing functions (to maintain communication with mobility) and location management functions (to keep track of the location of mobile nodes).
8. Limited power. Mobile nodes are battery operated and hence have limited power available. As a result, the power required for transmitting and receiving must be minimized in mobile wireless systems. Wireless Applications Wireless computing systems have several interesting and important applications such as: 1. Mobile e-commerce applications, popularly known as m-commerce applications. 2. Web surfing using wireless devices. 3. Access to corporate data by sales people and other frequent travelers while they are traveling. 4. Mobile video-on-demand applications. 5. Location-sensitive services, such as programs that help finding nearby movie theaters or restaurants an unfamiliar location. Wireless Communication Technologies Wireless communication systems use one or more of the wireless communication technologies discussed below Analog Cellular By using this communication technology, computing equipment that are equipped with a modem can transfer data over analog cellular telephone networks. Although the data rates supported by this technology are low (of the order of few Kbps), it is one of the most popular methods for wireless data communications due to the wide availability of modems and analog cellular services. Digital Cellular By using this communication technology, computing equipment can transfer digital data over digital cellular telephone networks without the need for a modem. The various standards available for digital cellular c transmission are 1. Time Division Multiple Access (TDMA), 2. Code Division Multiple Access (CDMA), and 3. Global System for Mobile Communication (GSM). The effective data transfer rate of digital cellular technology is less than that of analog cellular technology. Cellular Digital Packet Data (CDPD)
In this method, digital data to be transferred is broken into small packets and inserted into analog cellular telephone networks. The technology employs unused analog channels available in the analog telephone cellular infrastructure. That is, transmitters use a channel for a short transmission burst and release the channel for analog voice links. The effective data rate that can be achieved using this technique is 19.2 Kbps. Mobile Radio Networks The Specialized Mobile Radio (SMR) and Extended Specialized Mobile Radio (ESMR) technologies employ narrow bands for creating mobile radio network infrastructure. Such an infrastructure is primarily used for utility, dispatch, and other service-oriented metropolitan area users. Due to the use of narrow bands, mobile radio networks have limited data rates (of the order of 20 Kbps). Personal Communications Services (PCS) PCS technology is used to offer greater network complexity and lower end-to-end latency in wireless networks. It has two variations. The low-tier PCS is used for microcell coverage for frequency reuse and low-power benefits. On the other hand, the high-tier PCS is used for early digital cellular systems such as GSM, TDMA, and CDMA. It is used for large macrocells of the order of 0.5 and 3.0 kilometers. PCS technology may be used to serve LANs, MANs, and WANs. Satellite We have already seen that satellites are widely used for providing broadcasting, long distance, and international data communication services to stationary users. With advances in antenna design, signal reception, and other related technologies, it has now become possible to provide mobile services using satellites. Many satellite-based networks using Low-Earth-Oribit (LEO), Medium-Earth-Orbit (MEO), and Geosynchronous-Earth-Orbit (GEO) have been started to provide commercial wireless data services at relatively low data rates (below 64 Kbps). For example, Iridium is a LEO system that uses 66 satellites to provide mobile communications to every point on earth and within 50 miles above it. In Iridium, the same satellite may not serve a user throughout the duration of a call; a call may be handed off to another approaching satellite. The satellites, in concentric orbits, maintain links with up to four satellites in neighboring orbits. But handoffs are necessary among satellites in counter-rotating orbits to maintain cross-links among satellites. Although there is a large market for satellite-based mobile services especially in places where no (or poor) cellular/PCS service exists, due to the high initial cost
and per-minute charges, users of satellite-based mobile communications remain a small minority among a much larger number of wireless users. For example, Iridium, which started its service in January 1999, has been officially declared bankrupt and has terminated commercial services because the number of customers that adopted the service was much smaller than needed to maintain this multibillion-dollar network. Microwave Systems We have already seen that microwave systems use very high frequency radio signals to transmit data through space. Hence for several years, microwave systems have been and continue to be a popular means of wireless data communications. Microwave, links operate in higher frequency bands than any other technologies and offer data rates of several Mbps. Local Multipoint Distribution Systems (LMDS) LMDS is a technology for supporting point-to-multipoint applications. It uses a 28-31 GHz band with 1.3 GHz of spectrum allocated to it in the US and 3 GHz of spectrum allocated to it in Canada. LMDS uses spread spectrum and microcell technology that supports two-way data transfer rates of up to 1 Gbps. A drawback of the LMDS technology is that signal propagation at the 28 GHz range requires a direct line-of-sight path between the transmitter and receiver. Possible applications of LMDS technology are high-speed Internet data, video conferencing, and cable TV data transmission at several hundred Mbps. Wireless Technologies and Standards Over the years, several wireless technologies and standards have been introduced. Out of these, some of the most popular ones are described below. 2G and 3G Technologies The early mobile phones used first-generation technology, which was analog, circuit-based, narrowband, and suitable only for voice communications. Then came the second-generation wireless-networking technology (popularly known as 2G technology), which was digital, circuit-based, and narrowband but suitable for voice and limited data communications. Currently, we have the third-generation wireless-networking technology (popularly known as 3G technology), which is suitable for voice and advanced data applications, including online multimedia and mobile e-commerce. Further details of the 2G and 3G technologies are given below.
2G Technology As mentioned above, 2G technology is digital, circuit-based, narrowband, and suitable for voice and limited data communications. It comprises of the following three primary standards: 1. TDMA. The Time-Division-Multiple-Access (TDMA) technology does time-division multiplexing of each cellular channel to handle multiple communications simultaneously. In particular, TDMA technology divides each cellular channel into three time slots, each of which handles a separate transmission. The channel switches quickly from one time slot to another, thereby handling three communications simultaneously. This technology is primarily used in the US. 2. GSM. Global System for Mobile (GSM) communications technology is also based on time-division multiplexing of each cellular channel. However, it uses wider carrier frequencies than TDMA are divides each cellular channel into eight time slots, rather than three. Hence, GSM technology can handle eight communications simultaneously. This technology is primarily used in Europe and Asia. 3. CDMA. Code-Division-Multiple-Access (CDMA) technology does not divide a channel into sub-channels, like TDMA or GSM. Instead, it carries multiple transmissions simultaneously by filling the entire communications channel with data packets coded for various receiving devices. The packets go only to the devices for which they are coded. This technology is primarily used in the US. As upgrades to 2G technologies, 2.5G technologies were developed to provide more bandwidth for higher data transfer rates. 2.5G uses High-Speed-CircuitSwjtched-Data (HSCSD) technology, which is basically an extension of GSM technology that can offer throughput of up to 384 Kbps. Although 2.5G offers more bandwidth than 2G, it is less than that offered by 3G. However, since 2.5G uses existing 2G spectra and does not require: entirely new network infrastructure, it can be implemented faster and less expensively than 3G. 2.5G comprises of the following three primary standards: 1. iDEN. Integrated Digital Enhanced Network (iDEN) is a GSM upgrade that uses enhanced compression and modulation technologies to deliver data transfer rates of 64 Kbps. It is primarily used in North America, South America, China, and Japan. 2. GPRS. General Packet Radio Service (GPRS) is another GSM upgrade that offers data transfer rates of up to 171 Kbps. It can thus be used for applications such as videoconferencing and interacting with multimedia web sites. GPRS reallocates several GSM time slots for voice to data uses, thereby increasing data rates but decreasing voice rates. It is primarily used in China, Europe, and the US.
3. EDGE. Enhanced Data-rate for Global Evolution (EDGE) technology (also known as Enhanced Data GSM Environment) is yet another GSM upgrade that offers data transfer rates of up to 384 Kbps. It employs new modulation and radio techniques that use existing GSM's frequency range more efficiently. It is primarily used in tffe US. 3G Technology As mentioned before, 3G technology is designed to offer data rates high enough for mobile users to work with multimedia web content, videoconferencing, and ecommerce. It is designed to offer data transmission rates of up to 2.05 Mbps in stationary applications, 384 Kbps for slow-moving users, and 128 Kbps for users moving in vehicles. 3G is thus considerably faster than 2G and 2.5 G technologies. It comprises of the following three primary standards: 1. W-CDMA. Wideband Code-Division Multiple Access (W-CDMA) technology is an upgrade to the CDMA standard of 2G technology. It uses a 5MHz-wide carrier signal, rather than the 200-KHz-wide carrier that narrowband CDMA uses. The International Telecommunication Union (ITU) has formally standardized this technology to let the vendors offer a consistent set of services throughout the world. This technology is also known as International Mobile Telecommunications 2000 (IMT-2000) direct spread. 2. CDMA2000. This is yet another CDMA upgrade technology. It offers a migration path from CDMA2000 lx to CDMA2000 lxEV-DO and lxEV-DV to CDMA2000 3x. CDMA2000 lx offers maximum data rates of 153.6 Kbps. During the next few years, users could move to lxEV-DO (lx evolution, data only) and lxEV-DV (lx evolution, data and voice), offering data rates up to 614 Kbps. Finally, users could move ahead to CDMA2000 3x, which would provide the full 3G maximum throughput of 2.05 Mbps. 3. TD-CDMA. Time-Division CDMA (TD-CDMA) is a technology that combines time-division multiplexing and CDMA techniques. It was developed and used primarily in China. Wireless LANs Wireless Local Area Networks (WLANs) are like traditional LANs having a wireless interface to enable wireless communication among the equipment that are part of the LAN. A wireless LAN may be further connected to a more extensive fixed network such as LAN, WAN, the Internet, etc. The primary component of a wireless LAN is the wireless interface card that has an antenna. This interface card can be connected to the mobile unit as well as to the fixed network. Wireless LANs have limited range and are designed to be used only in local environments such as a building, hallway, park, or office complex. The main advantage is the
flexibility and mobility supported by a wireless LAN; bandwidth considerations are secondary. Unlike cellular networks with allocated channels (frequencies), users in WLANs have to share frequencies, which may eventually lead to collisions. The choice of frequency depends on whether microwave, spread-spectrum, or infrared communication will be used. Interference and security depend on the type of communications method used in the WLAN. The following three are the primary WLAN standards: 1. IEEE 802.11. The IEEE 802.11 standard supports 1 Mbps data rate and several choices of physical medium such as spread spectrum and infrared. It also supports prioritized access to the medium. An additional feature of this standard is battery conservation for inactive or idle wireless users. Many universities and companies are encouraging the use of IEEE 802.11-based LANs for accessing campus computing systems and the Internet. The IEEE 802.11 standard is being enhanced to support 11 Mbps data rate. 2. HiperLAN. The HiperLAN standard can be used to support 23.5 Mbps channel rates. It, however, allows the use of only spread spectrum physical medium. Like IEEE 802.11, it also supports prioritized access to the medium. 3. HiperLAN2. Another emerging WLAN standard is HiperLAN2, which is being standardized by ETSI (European Telecommunications Standard Institute). An exciting feature of this standard is that it provides for use of connections that offer different quality of service for different applications. It uses time-division multiplexing of unicast, multicast, and broadcast connections. Many major players in the WLAN area have formed HiperLAN2 Global Forum to advance and complement the ETS1 standardization process. Wireless Local Loops (WLLs) Wireless Local Loops (WLLs) technology has been designed to support fixed wireless systems. Since WLLs provide fixed wireless access (as opposed to mobile access provided by cellular and PCS), they can provide several MHz of bandwidth that can be used for high-speed Internet access and data transfer in addition to the basic telephone service. These loops allow long-distance carriers to bypass the existing wired local loops owned b) local telephone carriers (which may result in the saving of tremendous access charges). Due to this reason, man} long-distance companies in advanced countries (like the U.S.) are looking to build their WLLs to avoid paying access charges. Moreover, in developing nations where laying thousands of miles of copper cable is impractical WLLs can be used to provide telephone and low-speed data transfer services. Among many choices of technologies that can be used in WLL environments, cellular and microcellular system; can be used in the 900, 1800, and 1900 MHz
ranges with 9.6 Kbps for 10 to 100 users within a few kilometer; range. Wireless LANs can be deployed to support WLL for users in smaller areas but with higher bandwidth requirements. Systems that are especially designed for fixed wireless access, such as Local Multipoint Distribution Systems (LMDS), can provide very high bandwidth (tens of Mbps) in large areas for large number o users but require a direct line of sight. Radio-router Technology Radio-router technology is an emerging wireless technology designed to make links in an IP network mobile. It uses a radio-transmission framework for packetbased, broadband, IP wireless communications. A radio-router network can be built on top of the existing IP infrastructure, rather than from the ground up like a 3G network Since IP network technology is already well developed and inexpensive, radio-router systems will be relatively easy, quick, and economical to implement. The technology uses OFDM (Orthogonal Frequency Division Multiplexing), in which a single channel is divided into several sub-channels, each at a different frequency. This boosts bandwidth by letting a system carry several transmissions at the same time. Radio-router systems offer a maximum throughput of 1.5 Mbps. OFDM, unlike traditional FDM, uses signal modulation and demodulation techniques, as well as orthogonal placement of I adjacent channels, to minimize interference.
Multihop Wireless Networks In most wireless network environments, the radio channel is used to connect the mobile equipment to a base station in a single hop. The base station itself is connected to a wired infrastructure. Consider an environment f where there is no wired infrastructure yet connectivity must be maintained among users who can move around arbitrarily and can at times not be in direct transmission range of each other. Multihop wireless networks, also known as instant infrastructure systems, are designed to provide wireless communications capability in such environments. I Instant infrastructure systems play an important role in wireless communication because due to transmitted power constraints, not all radios are within the range of each other, and packets may need to be relayed from one radio to another on their way to their intended destinations. These systems, however, are more complex to design and manage than traditional single-hop, centrally controlled systems. This is mainly because a fully implemented multihop system needs to have complete support for multimedia, multihopping, and mobility features.
Wireless Application Protocol (WAP) We saw in the discussion above that many different wireless access technologies exist that are not interoperable. Designing and building networks and applications for each technology would certainly be a very difficult job for developers. With the proliferation of the Internet and wireless communications devices, Internet access using mobile devices has become important for anywhere, anytime information access capability. However, redesigning of all web sites to support downloading of information by mobile users is certainly not feasible. To address these issues, researchers started working towards the development of a technology that could allow development of applications that are independent of the underlying wireless access technology and that could also enable access of information from existing web sites using mobile devices. The result of this was WAP. WAP specifications have been developed by the WAP Forum, a consortium of leading wireless companies. It supports interoperability of different wireless networks, devices and applications using a common set of application and network protocols. In fact, WAP is a stack of protocols that defines the type and structure of messages that two devices have to use when they are communicating with each other in a wireless environment. The WAP protocol stack contains a set of six layers, which are application layer, session layer, transaction layer, security layer, transport layer, and wireless layer. Figure 17.26 presents a summary of the various WAP layers, the protocols at each layer, and their functions. WAP protocols have been designed to reduce the problems associated with the use of the Internet protocols for wireless data transfer. They do this simply by eliminating unnecessary data transfers and using binary code to reduce the amount of data that has to be sent. For instance, WAP has introduced a new Markup Language called the Wireless Markup Language (WML), which is a version of HTML optimized for mobile devices. Figure 17.27 shows the basic architecture of WAP. The architecture is similar to web architecture, but it requires a WAP gateway for translating between HTTP and WAP. That is, the WAP gateway acts as a proxy server to a mobile client and translates requests from WAP protocol stacks to protocol stacks employed by the information server on the other side. The WAP gateway has encoders to translate the content coming from the server into compact formats to reduce the size of data over the wireless network. The WAP gateway usually resides in the company or organization that provides carrier services to its subscribers. This architecture ensures that mobile users can access a wide variety of contents and applications and also allows application developers to build content services and applications that can run on a large base of mobile devices.
WAP Layers
Protocols
Functions
Application Layer
Wireless Application Provides an application environment required for Environment (WAE) the development and execution of the applications, that is, micro browser environment, WML, and WML script. Session Layer Wireless Session Supplies methods for the exchange of Protocol information between client and server (WSP) application. These methods include HTTP functions and semantics, facility for reliable and unreliable data push, and protocol feature negotiation. Transaction Wireless Transaction Provides several types of transaction services. Layer Protocol (WTP) Security Layer Wireless Transport Layer Provides authentication, privacy, and secure Security (WTLS) connection between applications. Transport Layer Wireless Datagram Provides a common interface to upper layer Protocol (WDP) protocols by adapting to specific features of the underlying technologies. Wireless Layer Wireless and mobile Provides a specific way to transmit information networks over a wireless link. Figure 17.26. The various WAP layers, the protocols at each layer, and their functions. DISTRIBUTED COMPUTING SYSTEMS The merging of computer and networking technologies gave birth to distributed computing systems in the late 1970s. Simply speaking, a distributed computing system is a collection of computers interconnected by a communication network. It is a configuration in which many independent computer systems (which may be geographically far apart) are connected by a communication network, and in which messages, processing task, programs, data, and other resources are transmitted between cooperating computer systems. Such an arrangement enables the sharing of many hardware and software resources as well as information among several users who may be sitting far away from each other. It also increases the usability of computers by bringing them closer to the end users and by integrating them into daily business activities at the locations at which these activities take place. The individual computers of a distributed computing system are often referred to as nodes.
It is obvious that distributed computing systems are much more complex and difficult to build than traditional centralized svstems. The increased complexity is mainly due to the following reasons: 1. The system must be designed in such a way so as to effectively use and manage a very large number of distributed resources. 2. The system must be designed to enable communication among the various nodes of the system. Special software is usually needed to handle loss of messages during transmission across the network or to prevent overloading of the network. 3. Special security measures are needed to protect the widely distributed shared resources and services against intentional or accidental violation of access control and privacy constraints. Despite the increased complexity and the difficulty of building distributed computing systems, the installation and use of distributed computing systems are rapidly increasing. This is mainly because the advantages of distributed computing systems outweigh their disadvantages. The major advantages that have led to the emergence and popularity of distributed computing systems are as follows [Reprinted, with permission,' from the book titled Distributed Operating Systems: Concepts and Design by Pradeep K. Sinha. © 1997 IEEE.]: 1. Inherently distributed applications. Several applications are inherently distributed in nature and require a distributed computing system for their realization. A few examples of inherently distributed applications are electronic mail facility, a computerized worldwide airline reservation system, a computerized banking system in which a customer can deposit/withdraw money from his or her account from any branch of the bank, and a factory automation system controlling robots and machines all along an assembly line. 2. Information sharing among distributed users. In a distributed computing system, information generated by one of the users can be easily and efficiently shared by the users working at other nodes of the system. For example, a project can be performed by two or more users who are geographically far off from each other but whose computers are a part of the same distributed computing system. In this case, although the users are geographically separated from each other, they can work in cooperation, for example, by transferring the files of the project, logging on to each other's remote computers to run programs, and exchanging messages by electronic mail to coordinate the work. The use of distributed computing systems by a group of users to work cooperatively is known as computer-supported cooperative working (CSCW), or groupware. Groupware applications depend heavily on the sharing of data objects between programs running on different nodes of a distributed computing system. 3. Resource sharing. Information is not the only thing that can be shared in a distributed computing system. Sharing of software resources such as software libraries and databases as well as hardware resources such as printers, hard disks,
and plotters can also be done in a very effective way among all the computers and the users of a single distributed computing system. 4. Better price-performance ratio. Distributed computing systems potentially have a much better price-performance ratio than a single large centralized system because they facilitate resource sharing among multiple computers. For example, a single unit of expensive peripheral devices such as color laser printers, high-speed storage devices, and plotters can be shared among all the computers of the same distributed computing system. If these computes are not linked together with a communication network, each computer must have its own peripherals, resulting in higher cost. 5. Shorter response times and higher throughput. Multiple processors of a distributed computing system can be utilized properly for providing shorter response times and higher throughput than a single-processor centralized system. For example, if a particular computation can be partitioned into a number of subcomputations that can run concurrently, in a distributed computing system all the subcomputations can be simultaneously run with each one on a different processor. Distributed computing systems with very fast communication networks are increasingly being used as parallel computers to solve single complex problems rapidly. Another method often used in distributed computing systems for achieving better overall performance is to distribute the load more evenly among the multiple processors by moving jobs from currently overloaded processors to lightly loaded ones. 6. Higher reliability. Reliability refers to the degree of tolerance against errors and component failures in a system. A reliable system prevents loss of information even in the event of component failures. The multiplicity of storage devices and processors in a distributed computing system allows the maintenance "of multiple copies of critical information within the system and the execution of important computations redundantly to protect them against catastrophic failures. With this approach, if one of the processors fails, the computation can be successfully completed at the other processor, and if one of the storage devices fails, the information can still be used from the other storage device. Furthermore, the geographical distribution of the processors and other resources in a distributed computing system limits the scope of failures caused by natural disasters. 7. Extensibility and incremental growth. Another major advantage of distributed computing systems is that they are capable of incremental growth. That is, it is possible to gradually extend the power and functionality of a distributed computing system by simply adding additional resources (both hardware and software) to the system as and when the need arises. For example, additional processors can be easily added to the system to handle the increased workload of an organization that might have resulted from its expansion. Incremental growth is a very attractive feature because for most existing and proposed applications it is practically impossible to predict future demands of the system. Extensibility is also easier in a distributed computing system because addition of new resources
.to an existing system can be performed without significant disruption of the normal functioning of the system. 8. Better flexibility in meeting users' needs. Different types of computers are usually more suitable for performing different types of computations. For example, computers with ordinary power are suitable for ordinary data processing jobs, whereas high-performance computers are more suitable for complex mathematical computations. In a centralized system, the users have to perform all types of computations on the only available computer. However, a distributed computing system may have a pool of different types of computers, in which case the most appropriate one can be selected for processing a user's job depending on the nature of the job. Points to Remember 1. A computer network is a network of geographically distributed multiple computers connected in such a way that meaningful transmission and exchange of information become possible among them. Sharing of information, sharing of resources (both hardware and software), and sharing of processing load are some of the major objectives of a computer network. 2. The three basic elements of any communication process are a sender (source), which creates the message to be transmitted, a medium, which carries the message, and a receiver (sink), which receives the message. 3. The three modes of transmitting data from one point to another are simplex, half-duplex, and full-duplex. In simplex mode of transmission, communication can take place in only one direction. A half-duplex system can transmit data in both directions, but only in one direction at a time. A full-duplex system allows information to flow simultaneously in both directions. 4. Bandwidth is the range of frequencies that is available for the transmission of data. The wider the bandwidth of a communication system, the more data it can transmit in a given period of time. 5. The communication data transfer rate is measured in a unit called baud. In general usage, baud is identical to bits per second. 6. There are several types of physical channels (communication media) through which data can be transmitted from one point to another. Some of the most common data transmission medium are wire pairs, coaxial cable, microwave system, communications satellite and optical fibers. -*'** . ' 7. The technique by which a digital signal is converted to its analog form is known as modulation. The reverse process, that is the conversion of analog signal to its digital form, at a destination device, is called demodulation. 8. The three forms of modulation are amplitude modulation, frequency modulation and phase modulation.
9. The process of modulation and demodulation, that is, the conversion of digital data to analog form and vice-versa, is carried out by a special device called a modem (modulator/demodulator). 10. Data transmission service providers are popularly known as common carriers. The various types of services offered by the common carriers are dial-up line, leased line, Integrated Services Digital Network (ISDN), and Value Added Network (VAN). 11. The method of dividing a physical channel into many logical channels, so that a number of independent signals may be simultaneously transmitted on it, is known as multiplexing. The electronic device that performs this task is known as a multiplexer. 12. The two basic methods of multiplexing channels are frequency-division multiplexing and time-division multiplexing. 13. The two modes of data transmission on a communication line are asynchronous and synchronous. In asynchronous transmission, the sender can send data at any convenient time and the receiver will accept it. On the other hand, in synchronous transmission, the sender and receiver must synchronize with each other to get ready for the data transmission before it takes place. 14. The three different methods of establishing communication links between the sender and receiver in a communication network are circuit switching, message switching and packet switching. 15. The selection of the actual path to be used to transmit a packet in a WAN is determined by the routing strategy used. 16. The term network topology refers to the way in which the nodes of a network are linked together. Although the number of possible network topologies is seemingly limitless, the four major ones are the star network, the ring network, the completely connected network, and the multi-access bus network. 17. Based on characteristics such as geographic distribution of nodes, data rate, error rate, and communication cost, networks are broadly classified into two types: LAN and WAN. Networks that share some of the characteristics of both LANs and WANs are sometimes referred to as MANs. 18. A communication or network protocol describes the rules that govern the transmission of data over communication networks. These rules provide a method for orderly and efficient exchange of data between the sender and the receiver and for the proper interpretation of controls and data transmitted as raw bits and bytes. 19. Network Interface card, often referred to as NIC or network card, is a hardware device that allows a computer to be connected to a network, both functionally and physically. 20. Computer networks are implemented using the concept of layered protocols. The OSI model provides a standard for layered protocols for WANs. The seven layers of the OSI model are physical, data-link, network, transport, session, presentation, and application. Of the available protocol suites for network systems, the Internet Protocol (IP) suite is the most popular and widely used.
21. Interconnecting of two or more networks to form a single network is called internetworking, and the resulting network is called an internetwork. The goal of internetworking is to hide the details of different physical networks, so that the resulting internetwork functions as a single coordinated unit. Tools such as bridges, routers, brouters, and gateways are used for internetworking. The Internet is the best example of an internetwork. 22. A wireless computing system enhances the functionality of computing equipment by freeing communication from the location constraints of the wired computing systems. 23. Wireless computing systems are of two types -fixed wireless systems that support little or no mobility of the computing equipment associated with the wireless network and mobile wireless systems that support mobility of the computing equipment which the users use to access the resources associated with the wireless network. 24. Wireless communication technologies used in wireless networks include analog cellular, digital cellular, Cellular Digital Packet Data (CDPD), mobile radio networks, Personal Communication Services (PCS), satellite, microwave systems, and Local Multipoint Distribution Systems (LMDS). 25. The commonly used wireless technologies include 2G and 3G technologies, wireless LANs, Wireless Local Loops (WLLs), radio-router technology, multihop wireless networks, and Wireless Application Protocol (WAP). 26. A distributed computing system is a configuration in which many independent computer systems (which may be geographically far apart) are connected by a communication network, and in which messages, processing task, programs, data, and other resources are transmitted between cooperating computer systems. Such an arrangement enables the sharing of many hardware and software resources as well as information among several users who may be sitting far away from each other. Questions 1. What is a computer network? How is it useful? 2. Identify the basic elements of a communication system and the purpose of each. 3. Differentiate between simplex, half-duplex, and full-duplex modes of data transmission. 4. Which mode of data transmission is suitable for communication between a terminal and a computer? 5. "A full-duplex line is faster since it avoids the delay that occurs in a halfduplex circuit." Explain this statement. 6. Explain the terms "bandwidth" and "baud". 7. Why are the transmission mode and the bandwidth of communication channels of concern to an analyst designing a data communication system?
8. Differentiate between narrowband, voiceband, and broadband communication channels. Give a practical application of each. 9. What is a wire pair? In what situations are they suitable for use in data transmission? 10. What is a coaxial cable? Give some of its practical uses. 11. Explain how microwave systems can be used for communication between two distant stations. 12. How are communications satellites used? What are the possible advantages and limitations of using a communications satellite? 13. What is an optical fiber? How is it used for data communications? What are its advantages? 14. Differentiate between analog and digital transmission of data. Give their relative advantages and disadvantages. 15. What do you understand by modulation and demodulation? 16. Why is modulation used in signal transmission? Describe the three different forms of modulation used in signal transmission. 17. What are modems? What purpose do they serve in data communication systems? 18. Describe some of the common factors to be considered while selecting a modem. 19. Explain the use of "fall back" feature in the design of modems. 20. Differentiate between internal and external modems. Give their relative advantages and disadvantages. 21. What is a FAX modem? How is it more useful than a conventional modem? 22. Describe the error correction and data compression features found in some modems. 23. What is an intelligent modem? List out some of the features found in intelligent modems. 24. What is a repeater? 25. When are public telephone lines used to connect a terminal to a computer? Are modems required for this purpose? Why? 26. Explain about the following types of services offered by the common carriers: (a) Dial-up line (c) Integrated Services Digital Network (ISDN) (b) Leased line (d) Value Added Network (VAN) 27. What is ISDN? Differentiate between narrow-band and broad-band ISDN. 28. What is a Value Added Network (VAN)? 29. What is a multiplexer? Explain its function with the help of a diagram. 30. Describe the two basic methods of multiplexing. Give uses of both the methods. 31. List out the differences between FDM and TDM. Which method is suitable for communication between computers and why? 32. What is a concentrator? Justify its use in a data communication system.
33. What is FEP? Illustrate its use. 34. Describe the asynchronous and synchronous modes of data transmission. 35. List out the relative advantages and disadvantages of asynchronous and synchronous modes of data transmission. 36. Explain how circuit switching is used to link the sender and the receiver in a communication network. What are the advantages and disadvantages of this method of switching? 37. Explain the store-and-forward method of message switching. Give the advantages and disadvantages of this message switching technique. 38. What is packet switching? Why is this method used for digital data communication between computers? 39. What is routing? What are the desirable features of a good routing algorithm? 40. Differentiate between source routing and hop-by-hop routing methods. Give their relative advantages and disadvantages. 41. Differentiate between static and dynamic routing methods. Give their relative advantages and disadvantages. 42. What is meant by network topology? Describe three commonly used network topologies with their relative advantages and disadvantages. 43. What is a hybrid network? Why are they used? 44. Write a short note on multi-access bus network. 45. What is a LAN? What are its main objectives? 46. What is WAN? What is a MAN? 47. Differentiate between a LAN and a WAN. Give one example of each. 48. What is a communication protocol? What are the normal functions performed by these protocols? 49. Why are communication protocols needed in a computer network? 50. What is a network interface card? Explain its usage in a computer system. 51. Why is layering used in the design of communication networks? 52. Describe the layering concepts in the OSI model of network architecture with the functions of each layer. 53. What is meant by internetworking? What are the main issues in internetworking? Explain the difference among the following terms: (a) Bridge (b) Router (c) Gateway 54. What is a wireless computing system? How is it useful? 55. Differentiate between fixed wireless systems and mobile wireless systems. 56. What are the major challenges in the successful use of wireless computing systems? 57. List out some of the applications that can benefit from wireless computing systems.
58. Describe the commonly used wireless communication technologies used in wireless communication systems. 59. Differentiate between 2G and 3G wireless technologies. 60. What is a wireless LAN? Describe the three primary WLAN standards. 61. What is a multihop wireless network? Why are such networks needed? 62. What is Wireless Application Protocol (WAP)? Why is it needed? 63. Describe the various WAP layers, the protocols at each layer, and their functions. 64. Describe the basic architecture of WAP? Explain the role of WAP gateway. 65. What is a distributed computing system? Why are distributed computing systems gaining popularity? 66. List out the advantages and limitations of a distributed computing system. 67. Write short notes on: (a) Data transmission modes (b) Data transmission speed 68. Write short notes on: (a) Microwave system (b) Communications satellite (c) Optical fibers 69. Write short notes on: (a) Network interface card (b) ISDN (c) Value Added Network (VAN) (c) Data transmission media (d) Data transmission services (d) Dial-up line (e) Leased line 70. Write short notes on: (a) Modem (b) Multiplexer (c) Concentrator (d) Front-end processor 71. Write short notes on: (a) FAX modems (b) Intelligent modems (c) Internal versus external modems (d) Transmission speeds of modems 72. Write short notes on: (a) Switching techniques (b) Routing techniques (c) Network topologies (d) Asynchronous and synchronous transmission 73. Write short notes on:
(a) (b) (c) 74. (a) (b) (c) 75. (a) (b) (c) (d) 76. (a) (b) (c) 77. (a) (b) (c) (d) (e) (f) (d) (e) (f)
LAN versus WAN Communication protocols Ethernet Write short notes on: The OSI model Internetworking tools Distributed computing systems Write short notes on: Fixed versus mobile wireless systems Issues in successful use of wireless computing systems Wireless applications Multihop wireless networks Write short notes on: Analog cellular Digital cellular Mobile radio networks Write short notes on: 2G technology 3G technology Wireless LANs Cellular Digital Packet Data (CDPD) Personal Communications Services (PCS) Local Multipoint Distribution Systems (LMDS) Wireless Local Loops (WLLs) Radio-router technology Wireless Application Protocol (WAP)1
Chapter 18 The Internet DEFINITION (WHAT IT IS?) The Internet is a huge network of computers that links many different types of computers all over the world. It is a network of networks that share a common mechanism for addressing (identifying) computers and a common set of communication protocols for communications between two computers on the network. BRIEF HISTORY The Internet has its root in the ARPANET system of the Advanced Research Project Agency of the U.S. Department of Defense. ARPANET was the first WAN and had only four sites in 1969. The Internet evolved from the basic ideas of ARPANET for interconnecting computers and was initially used by research organizations and universities to share and exchange information. In 1989, the U.S. Government lifted restrictions on the use of the Internet and allowed it to be used for commercial purposes as well. Since then, the Internet has rapidly grown to become the world's largest network. It now interconnects more than 30,000 networks, allowing more than 10 million computers and more than 50 million computer users in more than 150 countries around the world to communicate with each other. The Internet continues to grow at a rapid pace. ITS BASIC SERVICES There are basically four types of services provided by the Internet to its users. These are described below. Electronic Mail The electronic mail service (known as e-mail in short) allows an Internet user to send a mail (message) to another Internet user in any part of the world in a nearreal-time manner. The message may not reach its destination immediately, but may take anywhere from few seconds to several minutes, because it must be passed from one network to another until it reaches its destination. E-mail service has many similarities with the postal mail service that all of us are familiar with. All Internet users have an e-mail address, just like all of us have a postal address. Each Internet user has a logical mail box just like each one of us has a mail box in our house. When sending a mail to another user, the sender
specifies the e-mail address of the receiver just as we write the postal address of the receiver in the postal mail system. The e-mail service delivers an already sent mail into the receiver's mail box. The receiver extracts the mail from the mail box and reads it at his/her own convenient time just like in a postal mail system. After reading the message, the receiver can save it, delete it, pass it on to someone else, or respond by sending another message back. Messages in e-mail service can contain not only text documents but also image, audio and video data. The only restriction is that the data to be sent must be digitized, that is, converted to a computer-readable format. With e-mail service, the Internet has proved to be a rapid and productive communication tool for millions of users. As compared to paper mail, telephone, and fax, e-mail is preferred by many because of its following advantages: 1. It is faster than paper mail. 2. Unlike the telephone, the persons communicating with each other need not be available at the same time. 3. Unlike fax documents, e-mail documents can be stored in a computer and be easily edited using editing programs. File Transfer Protocol The File Transfer Protocol service (known as FTP in short) allows an Internet user to move a file from one computer to another on the Internet. A file may contain any type of digital information - text document, image, artwork, movie, sound, software, etc. Hence, anything that can be stored on a computer can be moved with FTP service. Moving a file from a remote computer to ones own computer is known as downloading the file and moving a file from ones own computer to a remote computer is known as uploading the file. By using the FTP service, a file transfer takes place in the following manner: 1. A user executes the ftp command on his/her local computer, specifying the address of the remote computer as a parameter. 2. An FTP process running on the user's computer (called FTP client process) establishes a connection with an FTP process running on the remote computer (called FTP server process). 3. The user is then prompted for login name and password to ensure that the user is allowed to access the remote computer. 4. After successful login, the desired file(s) are downloaded or uploaded by using get (for downloading) and put (for uploading) commands. The user can also list directories or move between directories of the remote computer before deciding the file(s) to be transferred.
Notice from the discussion presented above that a user could transfer files to/from a remote computer only if he/she has access rights for the remote computer. With this restriction, it is almost impossible to provide access rights to the vast number of the users of the Internet to a computer that contains sharable information. The concept of anonymous FTP site is used to overcome this problem. An anonymous FTP site is a computer that allows a user to log in with a username of anonymous and then a password that is the user's e-mail address. The user can then download files from the computer that have been stored on it for the purpose of sharing. Such sites are called publicly accessible sites because they can be accessed by any user on the Internet. Of course, FTP service is also used for more secure file transfer operations. In such cases, a user needs a valid username and password to access a particular computer. This is common; for example, for organizations that wish to let only certain people access their computer. Telnet The telnet service allows an Internet user to log in to another computer somewhere on the Internet. That is, a user can execute the telnet command on his/her local computer to start a login session on a remote computer. This action is also called "remote login." To start a remote login session, the user types the command telnet and the address of the remote computer on the terminal of his/her local computer. The user then receives a prompt asking to enter a login name (user ID) and a password to ensure that the user has the access rights for accessing the remote computer. If the user specifies a correct login name and password, he/she gets logged in to the remote computer. Once the login session is established with the remote computer, telnet enters the input mode and anything typed on the terminal of the local computer by the user is sent to the remote computer for processing. Some of the common uses of the telnet service are: 1. Using the computing power of the remote computer. The local computer may be an ordinary personal computer and the remote computer may be a powerful super computer. 2. Using a software on the remote computer. The software that the user wants to use may not be available on the local computer. 3. Accessing remote computer's database or archive. The information archive such as the public database or library resources may be available on the remote computer. 4. For logging in to ones own computer from another computer. For example, if the user is attending a conference in another city and has access to a computer on
the Internet, he/she can telnet to his/her own computer and read his/her electronic mails. Usenet News The usenet service allows a group of Internet users to exchange their views/ideas/information on some common topic that is of interest to all the members belonging to the group. Several such groups exist on the Internet and are called newsgroups. For example, the newsgroup named comp. security misc consists of users having interest in computer security issues. A newsgroup is like a large notice board that is accessible to all the members belonging to the group. A member, who wants to exchange his/her views/ideas/information with other members, sends a specially formatted message, which is processed and interpreted as a notice to be posted on the notice board by the member's computer, which has an appropriate software running for this purpose. After the processing of the message is over, it can be read (seen) from any member's computer just like a notice posted on a notice board can be read by anyone having access to the notice board. There are two types of newsgroups - moderated and nonmoderated. In a moderated newsgroup only selected members have the right to directly post (write) a message to the virtual notice board, whereas in a nonmoderated newsgroup any member can directly post a message to the virtual notice board. In a moderated newsgroup, a message sent by an ordinary member first goes to the computer of one of the selected members having the right to directly post a message to the notice board. The message is then checked by this member who decides whether the message is appropriate for being posted and does any editing that might be required to improve the quality of the message. This member then posts the message on the notice board. After this only the message becomes readable to (can be seen by) other members. The main idea of a moderated newsgroup is to ensure the quality of the newsgroup. As no one moderates the quality of the messages of a nonmoderated newsgroup, anyone can post anything. But members of such newsgroups should be careful in posting a message because other members may react angrily by sending a flaming message in response to an ill mannered message. To address this issue, the concept of netiquette (network etiquette) has arisen, that deals with rules of framing messages that will not hurt others. Each newsgroup has a unique name. For example, the newsgroup named comp.lang.c contains messages relating to the discussion on C programming language issues. A newsgroup can be created by the system administrator of any
computer on the Internet. To make it accessible to other members, he/she then exports the newsgroup to other computers by using a special command meant for this purpose. An Internet user who wants to become a member of a newsgroup has to first subscribe to the newsgroup by using a special command meant for this purpose. An already subscribed user can unsubscribe from a newsgroup at any time. Over the years, many newsgroups have become extremely well organized and moderated and include valuable information. For example, many newsgroups have documents called FAQs (Frequently Asked Questions) that try to answer many of the questions that someone new to the group would have on the topic. A newsgroup is a great place to ask questions or catch up on the recent issues of the topic it covers because newsgroups often have world experts in their respective fields as their members. INTERNET SEARCH TOOLS The Internet is a vast ocean of information that is of interest to a wide variety of users. Obviously, this large information bank will be of little or no use if one can't easily find something of his/her own interest in it. Therefore, several user-friendly tools have been made available that allow users to successfully navigate the Internet and find useful information for one's own purposes. A few of the most popular of these tools are briefly described below. The process of navigating the Internet to search for useful information is often called Internet surfing. Hence, Internet search tools are also called surfing tools. Gopher Gopher is a text-based tool that provides hierarchical collections of information of all sorts across the Internet. It displays the information hierarchy as a series of menu items. That is, when using gopher, each menu item connects to another menu that branches off to yet another series of menus until you reach the actual document, image, sound, file, etc. Gopher is useful for an Internet user who is faced with the question "Is there anything on the Internet that is useful for me?" It allows the user to see information on the Internet as if they were all on his/her own computer. That is, from the menus displayed by gopher, the user can select and display the desired information without having to type the address of the computer on which the information is located or without having to remotely log in to that computer. Gopher automatically does all this for the user in the background.
Archie Archie is a keyword-based search tool that allows files, stored at anonymous sites of the Internet, to be searched either by their names or by their contents. You tell archie to search for something by typing a "keyword". In case of a "name search", the keyword is a filename or part of a filename you are looking for. As a result of this search, archie will display a listing of FTP site addresses that have filenames that contain the indicated keyword. The filenames containing the indicated keyword are also displayed for each such FTP site. On the other hand, in case of a "content search", the keyword is a word or phrase that you are looking for as data in files. As a result of this search, archie will display a listing of FTP site addresses and the filenames of the files that contain the indicated keyword as file data. World Wide Web The World Wide Web (called WWW or W3 in short) is the most popular and promising method of accessing the Internet. The main reason for its popularity is the use of a concept called hypertext. Hypertext is a new way of information storage and retrieval that enables authors to structure information in novel ways. An effectively designed hypertext document can help users rapidly locate the desired type of information from the vast amount of information on the Internet. Hypertext documents enable this by using a series of links. A link can be shown on the screen in multiple ways such as a labeled button, highlighted text, or different color text than normal text if your computer has a color display, or author-defined graphic symbols. A link is a special type of item in a hypertext document that connects the document to another document that provides more information about the linked item. The latter document can be anywhere on the Internet (in the same document in which the linked item is, in the same computer in which the former document is, or in another computer that is at the other end of the world). By "connect" we mean that a user simply selects the linked item (using a mouse or key command) and the user almost immediately sees the other document on his/her computer terminal. The concept of hypertext can be best illustrated with the help of an example, Suppose the screen of your computer terminal has the following hypertext document currently displayed on it: Pradeep K. Sinha has been involved in the research and development of distributed systems for almost a decade. At present Dr. Sinha is working at the Centre for Development of Advanced Computing (C-DAC), Pune, India. Before joining C-DAC, Dr. Sinha worked with the Multimedia Systems Research Laboratory (MSRL) of Panasonic in Tokyo, Japan.
This hypertext document has the following two links that are shown on the screen as highlighted (bold and underlined) texts: 1. Centre for Development of Advanced Computing (C-DAC). Suppose this link connects the current document to another document that gives detailed information about C-DAC and is located on a computer system at C-DAC in Pune, India. 2. Multimedia Systems Research Laboratory (MSRL) of Panasonic. Suppose this link connects the current document to another document that gives detailed information about MSRL of Panasonic and is located on a computer system at MSRL of Panasonic in Tokyo, Japan. Now, suppose your computer had a mouse connected to it, and you used the mouse to click anywhere on the link Multimedia Systems Research Laboratory (MSRL) of Panasonic of the displayed document. What would happen? Within a few seconds you will find yourself connected to the computer at MSRL of Panasonic in Tokyo and displayed on your computer screen will be the document that gives detailed information about MSRL of Panasonic. Hypertext documents on the Internet are known as Web Pages. Web Pages are created by using a special language called HyperText Markup Language (HTML in short). HTML is a subset of the more generalized language called Standard Generalized Markup Language (SGML in short) which is a powerful language for linking documents for easier electronic access and manipulation of documents. HTML is becoming a de-facto industrial standard for creating Web Pages. The WWW uses the client-server model and an Internet Protocol called HyperText Transport Protocol (HTTP in short) for interaction between the computers on the Internet. Any computer on the Internet that uses the HTTP protocol is called a Web Server and any computer that can access that server is called a Web Client. The use of the client-server model and the HTTP allows different kinds of computers on the Internet to interact with each other. For example, a Unix workstation may be the web server and a Windows PC may be the web client if both of them use the HTTP protocol for transmitting and receiving information. WWW BROWSERS To be used as a web client, a computer needs to be loaded with a special software tool that is known as WWW browser (or browser in short). Browsers normally provide the following navigation facilities to help users save time when they are jumping from server to server while Internet surfing:
1. Unlike FTP and Telnet, browsers do not require a user to remotely log in to a server computer and then to log out again when the user has finished accessing information stored on the server computer. 2. Browsers allow a user to specify an URL address of a server computer to facilitate the user to directly visit the server computer's site and to access information stored on it. URL stands for Uniform Resource Locator. It is an addressing scheme used by WWW browsers to locate sites on the Internet. 3. Browsers allow a user to create and maintain a personal hotlist of favorite URL addresses of server computers that the user is likely to frequently visit in future. A user's hotlist is stored on his/her local web client computer. Browsers provide hotlist commands to allow the user to add, delete, update URL addresses in the hotlist and to select an URL address of a server computer from the hotlist when the user wants to visit that server computer. 4. Many browsers have a "history" feature. These browsers maintain a history of the server computers visited in a surfing session. That is, they save (cache) in the local computer's memory, the URL addresses of the server computers visited during a surfing session, so that if the user wants to go back to an already visited server later on (in the same surfing session), the link is still available in the local computer's memory. 5. Browsers allow a user to download (copy from a server computer to the local computer's hard disk) information in various formats (i.e., as a text file, as an HTML file, or as a PostScript file). The downloaded information can be later (not necessarily in the same surfing session) used by the user. For example, downloaded information saved, as a PostScript file can be later printed on a PostScript-compatible printer, where even the graphics will be properly reproduced. Browsers come in various forms and capabilities. The currently available browsers can be broadly classified into the following three types: 1. Line browsers, 2. Graphical browsers, and 3. Java-enabled browsers. They are described below. Line Browsers A line browser is a text-based browser that provides command-line interface to its users. That is, line browsers are capable of displaying only plain text. They denote a hypertext link either by using numbers in brackets like this [1], or by using boldface type like this. Line browsers that use the latter approach to denote a hypertext link are known as Lynx browser and are the most common line browsers in use today. Line browsers are used on those web-client computers that only have text displaying capability.
An example of a typical line browser display is given in Figure 18.1. Note that every word or phrase that is of boldface type (such as ARPANET system) represents a hypertext link. The commands at the bottom of the screen help users take advantage of the various features of the line browser. For example, a user can use the up and down arrow keys to skip to the next hypertext link, and use the right and left arrow keys to either select a link, or move backward to a previous link. Pressing the return key initiates the link that is currently selected and the browser takes the user to wherever it leads.
An Introduction to the Internet The Internet is a network of networks that share a common mechanism for addressing computers and a common set of communication protocols for communications between two computers on the network. It has its root in the ARPANET system. The following are the four types of basic services provided by the Internet to its users: Electronic Mail File Transfer Protocol Telnet Usenet News There are several user-friendly, tools that allow users to successfully navigate the Internet and find useful information for one's own purposes. A few of the most popular of these tools are: Gopher Archie - World Wide Web To be used as a web client, a computer needs to be loaded with a special software tool that is known as WWW browser. — press space for next page — Arrow keys: Up and Down to move. Right to follow a link; Left to go back. H>elp 0>ptions P>rint G>o M>ain screen Q>uit/=search [delete] = history list
0 Figure 18.1. A typical line browser display. Graphical Browsers Although line browsers offer full hypertext capability for Internet surfing, they have the following limitations:
1. They cannot handle multimedia data such as graphics, image, video, and sound. 2. They rely heavily on the keyboard for navigating and selecting information because a user has enter a selection number or skip from link to link by using the arrow keys of the keyboard. Graphical browsers are designed to overcome these limitations of line browsers. They are designed to take full advantage of the WWW's ability to handle all kinds of digital media such as graphics, image, video and sound. Moreover, they rely heavily on the mouse for navigating and selecting information. A user can literally point a: click his/her way around the world on the Internet by using only the mouse. However, graphical browsers requi the web client computer to have graphics capabilities of handling and displaying multimedia data. Therefore, i web client computer has graphics capabilities, such as a Windows, Macintosh, or XWindows machine, then it c be loaded with a graphical browser to allow its users to have a visually more spectacular Internet surfing compared to a line browser. These are many graphical browsers available today with names like Mosaic, NetScape, Cello, Viola, MacWeb, Internet Explorer, etc. Out of these, Mosaic is the best known and probably the most popular one. It is distributed free of charge by the NCSA (National Center for Supercomputing Applications). There is also a commercial version of Mosaic available now called NetScape, which is becoming the de-facto standard for browsers. Internet Explorer is also a very popular graphical browser. To give an idea of a graphical browser, Figure.,18.2 shows how the initial screen looks like when Internet Explorer is invoked. Note that in case of a graphical browser, a hypertext link can be shown on the screen in multiple ways such as a labeled button, highlighted (bold and/or underlined) text, or different color text than normal text if the computer has a color display, or author-defined graphics symbols. If the computer has a color display, the usual default is for links to be in both blue and red colors. Blue links are places you haven't visited yet in your current Internet surfing session and Red links are places you have already visited. In other words, once you click on a link and go to that place, if during the same session you ever come back to the text from which the hypertext jump was made, it will show up as a red link indicating that you have already visited the place of this link. This little navigation helper is a great time-saver if you are jumping from server to server searching for information on a particular topic. Java-enabled Browsers A conventional graphical browser works as follows (see Figure 18.3):
1. The user of the web-client calls on hypertext link. 2. The browser of the web-client computer decodes the URL address of the link. 3. The browser contacts the server corresponding to the URL address and requests for the object referenced by the link. 4. The server then fetches an object from the local computer and returns to the browser a bag of bytes that contains the object. 5. The browser then inspects the bytes to determine the type of the data such as HTML document or JPEG image, etc. 6. Finally, the browser uses the data type information to invoke the appropriate code to manipulate and display the object on the web-client computer's screen. Browser - User clicks on a hypertext link Browser - Browser decodes URL address of the link Browser - Browser requests corresponding object from the server Browser - Browser inspects the bytes to determine the type of the data Browser uses the data type information to invoke the appropriate code to manipulate and display the object on the computer's screen Network - request Network - Reply Server - Server fetches the object and returns to the browser a bag of bytes that contains the object Conventional graphical browsers that work in the manner described above are called "first-generation" browsers. The main problem with these browsers is that they are built in a monolithic fashion. That is, every possible type of data, protocol and behavior that a browser can deal with is frozen into its design when the browser is built. This means that every time a new data type, protocol, or behavior is invented, this browser must be upgraded to make it capable of dealing with the new situation. From the viewpoint of end users this is a difficult job because users must continually be aware of what data types and protocols exist, which browsers deal with which of these data types and protocols, and which versions of which browsers are compatible with each other. Given the rapid growth of the Internet, this situation is clearly out of control. Java-enabled browsers, known as "secondgeneration" browsers, are designed to overcome the above-mentioned limitations of the conventional graphical browsers. They are based on the Java programming language (see Chapter 12 for details) and have the ability to dynamically download Java applets (mini-applications) from webserver computers to webclient computers and execute them on the client's computer. This feature provides the following capabilities to Java-enabled browsers: 1. Dynamic content. We saw that a first-generation browser can download data from a server computer to its own client computer and can display the data on the client's computer terminal screen. In addition to data, Java-enabled browsers can also download Java-based applets (segments of programs that act as miniapplications) on the client comnnter that are executed right there on the client
machine. Java-based applets may be written to pntvid" full animation, play sound, and interact with the user in real time when executed. Thus, Jav3 ^n^iled browsers transform the static data display capability of first-generation browsers into a new td dynamic viewing system. They enable creation and display of animation-oriented applications that was not possible with first-generation browsers. For example, someone could write a Java program to implement on interactive chemistry simulation. Users browsing the Internet with a Java-enabled browser could easily download this simulation program on their computer and interact with it, rather than just having a static picture with some text displayed on their computer screen. 2. Dynamic data types. We saw that due to their monolithic design structure, the first-generation browsers can deal with only a fixed set of data objects. For example, a browser might be designed to understand only the following image formats - GIF, XI1 pixmap, and XI1 bitmap. If this browser sees a new image format, it cannot display it because it does not know how to deal with it. A Javaenabled browser overcomes this problem of dealing with an unknown/new data type object because it can dynamically download the Java-based applet for dealing with the unknown data type object from the server computer from which the data object is downloaded. The applet can then be executed to deal with the unknown data type object and to display it on the client's computer screen. Figure 18.4. illustrates how a Java-enabled browser interacts with a server when it encounters an object of an unknown type. Notice that the capability of Java-enabled browsers to understand different types of objects dynamically makes life of the Internet users much more comfortable. For example, if someone invents a new object type, the inventor just has to make sure that a copy of the Java-based applet for dealing with the object type is installed on the server that contains the object type. There is no need to upgrade all the browsers in the world. A Java-enabled browser actually upgrades itself dynamically when it encounters a new type of object. 3. Dynamic protocols. HTTP (HyperText Transport Protocol) is the most commonly used protocol for interaction between the computers on the Internet. Hence, most of the first-generation browsers have the knowledge of HTTP built in them. However, with the increase in commercial use of the Internet, many software vendors are providing new browsers and servers with added capabilities, such as billing and security. These new capabilities most often take the form of new protocols. So each vendor comes up with their unique style of security and sells a server and browser that understand this new protocol. An increasing trend of such new protocols have become a problem for Internet users because if a user wants to access data on multiple servers with each one having its own proprietary new protocols, the user needs multiple browsers or needs to upgrade its browser every time a new protocol is introduced. This is certainly not possible. ' A Java-enabled browser overcomes this problem of dealing with new protocols because it can dynamically download the Java-based applet for dealing with the
new protocol from the server computer with which it is interacting. The software vendor of a server with a new protocol just has to make sure that a copy of the Java-based applet for dealing with the new protocol is installed on the server. The interactions between the browser and server take place in a manner similar to the one illustrated in Figure 18.4. A Java-enabled browser actually upgrades its capability of understanding new protocols dynamically as and when it encounters a new protocol. Browser User clicks on a hypertext link Browser decodes URL address of the link Browser requests corresponding object from the server Browser inspects the bytes to determine the type of the object and realizes that it is of unknown type Browser requests the server for the Java-based applet to support the object Browser executes the applet to manipulate and display the object on the computer's screen Network Request Reply Request Reply Server Server fetches the object and returns to the browser a bag of bytes that contains the object Server returns to the browser the applet to support the object Figure 18.4. Functional structure of a Java-enabled browser. INTERNET SEARCH ENGINES An Internet search engine is an application available on the WWW that helps users locate the web sites containing useful information and references to such information. For using a search engine, the user has to type the description of the desired information using the user interface provided by the search engine. The search engine in turn searches for the requested information and returns the results of this search to the user. The returned results enable the user to quickly locate the requested information from the vast ocean of information available on the Internet.
Major Elements of Internet Search Engines The three major elements of an Internet search engine are as follows: 1. Search request interface, which allows the users to provide description of the desired information to the search engine. Depending on the complexity of this element, the search engine may allow specifications of simple keywords, combination of keywords using boolean operators and exclusion/inclusion operators, specification of phrases, and title and URL limiters. 2. Information discoverer, which discovers information from the WWW and creates a database for the search engine, which is used to locate useful information during the search process. The information discoverer accrues this information in two-ways: (a) in the manual method, authors provide information about their websites to the search engine; (b) in the automatic method, the information discoverer collects the information using programs, such as web crawlers, spiders, robots. These programs periodically roam on the Internet storing links of information about each page they visit as well as their associated hyperlinks. The information discoverer creates an index from the results of the automated search. If the web page is changed or updated, the spider finds these changes and the index is updated. 3. Presenter of search results, which returns the results of the search, ranking them in the order it believes, is most relevant to the search. Search engines often list the results of the search in accordance to a relevance score. Relevance scores reflect the number of times a search term appears in a web page. Some search engines also allow the user to control the relevance score by giving different weights to each search word. Categories of Internet Search Engines Internet search engines are normally categorized into the following four categories: 1. Natural Language Search Engines. These search engines allow the users to submit search phrases in natural languages such as English, Hindi, Japanese, etc. instead of using boolean operators to link keywords. , 2. Subject Directories Search Engines. These search engines present the results of searches in hierarchical subject categories, instead of listing them in an order by how relevant the search engine thinks the site is to the search. 3. Subject Guide Search Engines. These search engines contain links of many important resources on a topic. They differ from subject directories search engines in the sense that their search request interface has forms that contain databases of sites on many subjects, organized into a subject classification scheme. 4. Meta Search Engines. These search engines do not create their own databases but make use of the databases of other search engines. They quickly process the
user's search request by invoking several individual search engines at once and return the results compiled into a convenient format. Some Popular Internet Search Engines Today, there are more than two dozen major search engines available on the WWW. Some of the popular ones are as follows: 1. HotBot (www.hotbot.com) provides extensive search options through the use of query operators and forms. It allows the use of query operators in combination with forms, which many other search engines do not allow. In addition to the options found on the basic search form, it has an advanced search form.that provides more precise filters, additional specifiers, and limiters. 2. Yahoo (www.yahoo.com) supports a query syntax that allows inclusion/exclusion, phrase matching, wildcard matching, and title and URL limiters. Like other search engines, Yahoo provides form-based searching on its search options page, which also allows customization of the results display, and date ranges. 3. Lycos (www.lycos.com) provides a good option of both basic and advanced search features. Its basic search form has a drop-down menu that allows boolean and phrase searching, natural language queries, and basic adjacency and proximity searches. On the other hand, it has advanced search options, which allows the user to limit results to sound or picture files, or pages offered as a part of the Lycos services. These limiters may also include HTML title, URL, and specific domains. 4. Infoseek (www.infoseek.com) uses a form-based approach that requires a Java script enabled browser. It assumes some knowledge of HTML to provide the description of the desired information. 5. Google (www.google.com) is an interesting search engine having many unique features. It excels in producing extremely relevant results, and instead of web page summaries shows an excerpt of the text that matches the search query. For determining relevance scores of search results, it goes far beyond the number of times a keyword appears on a page and examines all aspects of the page's contents. It prioritizes the results according to how closely the individual search terms appear, and favors results that have the search terms near each other. It only produces results that match to all the search terms either in the text of the page or in the text of the links pointing to the page. 6. Inference Find (www.infind.com) is a meta search engine that searches six search engine: simultaneously. The search engine merges the results, removes duplicate files, and organizes the results into sections by concepts and/or by toplevel domain (educational institution, government organization, company, etc.) It retrieves the maximum number of results each search engine will allow by searching target search engines in parallel. Due to the parallel search feature, it returns the results very quickly.
7. Ixquick (www.ixquick.com) is another meta search engine that uses multiple search engines to perform the search and returns only the top-10 of the search results. It utilizes a "star" system, whereby the number of stars indicates the number of sites ranking each result in the top-ten. It offers a variety of search options including boolean, implied boolean, natural language search, truncation, case sensitivity and field search by sending a query to the search engines that support these options. It also has features for searching news, mp3 music files and pictures. USES OF THE INTERNET The world-wide scope of the Internet makes it perhaps the single most valuable tool for use in many significant ways by both non-profit and commercial organizations. Some of the important current strategic uses of the Internet are listed below: 1. On-line communication. The electronic mail service on the Internet is extensively used today by computer users around the world to communicate with each other. With this facility, the Internet has proved to be a rapid and productive communication tool for millions of users. 2. Software sharing. The Internet provides access to a large number of shareware software development tools and utilities. A few examples of such available shareware tools are compilers, code libraries, mail servers, and operating systems. For example, a set of such shareware tools is available via ftp from the Internet site sunsite.unc.edu. The Free Software Foundation also provides a wealth of GNU software tools on the Internet that can be downloaded free of charge. For more details on the software tools offered by the Free Software Foundation, you can anonymous yip to the Internet site prep.at.mit.edu and look in the directory /pub/GNU. 3. Exchange of views on topics of common interest. The Internet has a number of news groups. Each news group allows a group of users to exchange their views on some topic of common interest. For example, the news group comp.os.os2.advocacy contains candid dialog about the OS/2 operating system. 4. Posting of information of general interest. The Internet is also being extensively used as a large electronic bulletin board on which information of general interest can be posted to bring it to the attention of interested users around the world. Some commonly posted information includes career opportunities, conference and event announcements, and calls for papers for conferences and journals. 5. Product promotion. Several commercial organizations are effectively using the Internet services for promoting their products. These organizations make use of corporate ftp, Gopher, or WWW server sites focused on disseminating timely information about corporate happenings, product announcements, recent strategic alliances, press releases, and other information of potential interest
to existing and prospective customers. For example, comp.sys.sun.announce news group contains information about Sun Microsystem's latest product announcements. 6. Feedback about products. In addition to product promotion, commercial organizations are also using the Internet to gather information about user satisfaction of existing products, market opportunities of new products, and ideas for potential new products. This is usually accomplished by putting up an interactive survey application by the organization on a WWW or Gopher site on the Internet. 7. Customer support service. Many software organizations are also using the Internet to provide unprecedented levels of timely customer support. The combined electronic mail, ftp, and other services on the Internet provide all of the enabling tools necessary to provide such first-rate customer support. For example, bugs in fielded software products can be reported to an organization via electronic mail, and bug fixes, minor releases, workarounds, known problems and limitations, and general advice about a product can be made available by an organization to its customers via an ftp server. 8. On-line journals and magazines. The Internet now has literally thousands of electronic subscriptions that can be found both for free and low cost. There are many Gopher and WWW sites on the Internet that deal with electronic versions of many journals and magazines. For example, the Internet edition of the Times of India newspaper is available at the Internet site http://www.timesofindia.com. Researchers are working in the direction to extend this idea to support full-fledged electronic libraries on the Internet. 9. On-line shopping. The Internet has also facilitated the introduction of a new market concept that consists of virtual shops. These shops remain open 24 hours all the year round and are accessible to purchasers all around the world. They provide information about products or services for sale through ftp, Gopher, or WWW servers. Using the Internet services, customers submit specific product queries and request specific sales quotes. Through a well-defined authorization and authentication scheme, the Internet services are then used to accept orders placed by the customers, to handle order payments, and to track orders to fulfillment. For example, the Internet Mall is a collection of shops, each providing several products or services for sale. For a list of the available products or services at the Internet Mall, you can ftp to the Internet site ftp.netcom.com and look in the directory /pubs/Guides. Similarly, the Internet site amazon.com is a WWW-based book shop on the Internet where information on all types of International books can be found and books can also be ordered on-line. 10. World-wide video conferencing. World-wide video conferencing is an emerging service on the Internet that allows a group of users located around the globe to talk and interact with each other as if they were sitting and discussing in a single room. The parties interacting can see each other talking on their computer screens and can hear each other's voice through a special audio-device fixed in
their computers. The CU-SeeMe system developed at Cornell University is an example of an Internet-based video conferencing system. For information on CUSeeMe, you can ftp to the Internet site gated.cornell.edu and look in the directory /pub/video/CU-SeeMe.FAQ. 76.txt. HOW TO GET CONNECTED TO THE INTERNET? We saw that the Internet can be put to many different types of usage. However, before one can even think of using the Internet, he/she has to be connected to this worldwide communications network. There are four basic ways to get connected to the Internet: 1. Direct/Leased-line 2. Remote Dial-up 3. SLIP/PPP . 4. Commercial Online Service Each of these options varies in terms of hardware and software required for connection, cost of connection, and flexibility of usage (what can and cannot be done). Let us consider these options one by one to understand their connectivity differences. Direct/Leased-line Connection In this method, a dedicated (leased) phone line is used to connect a computer system to the Internet. As the communication line is dedicated, the entire bandwidth of the line is reserved for network traffic between the Internet and the connected computer system. The connected computer system is sometimes called a gateway because it can be used to provide Internet connectivity to other computers via it. Here direct connection means that the connected computer system forms a domain (a final destination) on the Internet, such as stanford.edu or ibm.com, and that it can be used to provide Internet connection to other computers that become part of the same domain. Most large organizations having their own internal computer network (such as colleges, universities, research institutions, etc.) use this method of Internet connectivity. Depending on the type of setup within an organization, individual computers of the organization may or may not have their own Internet address and software tools for communicating with the Internet. In the former case, access to the Internet from an individual computer is direct, meaning that a user of the individual computer can use its own software tools for Internet access and if he/she transfers a file from another computer on the Internet, that file will be downloaded directly to his/her computer's hard disk, where he/she can immediately access it. In the latter case, the user of an individual computer has to use the software tools for Internet access that are on the gateway computer and a
file downloaded from the Internet first goes to the gateway computer's hard disk which has to be again transferred to the hard disk of the user's computer. Fees for direct connections are often based on yearly contracts and the cost of such a connection depends on a number of factors, including bandwidth (speed of the leased line), number of users, and the purpose of use (for profit, non-profit, educational, etc.). Because of the significant expenses involved, direct connection method is suitable mainly for large organizations and software companies and is not suitable for individuals willing to connect their own personal computer or workstation to the Internet. Remote Dial-up Connection We saw that the gateway computer system of a direct/leased-line connection can be used to provide Internet connectivity to other computers via it. This idea has led to the concept of Internet Service Provider (ISP). An ISP is an organization that maintains a gateway computer system having direct/leased-line connectivity to the Internet and provides Internet access facility to other users (that are registered with it) by allowing them to connect their personal computers to its own gateway computer by using a modem. This method of connecting a user's personal computer to the Internet is known as remote dial-up connection. The ISP's gateway computer is called a host or server, and the user's computer is called a client or remote machine. Using his/her personal computer, the user has to first log in to the server computer in order to use its connection to the Internet. rFor a remote dial-up connection, a user requires a computer with some basic telecommunications software, a • telephone line, and a modem. Today, you can buy a computer with a modem already installed in it. However, if fyou already have a computer and you need to buy only a modem, the recommenced speed for a new modem is 128.8 Kilobits/sec or faster. With this kind of connection, a user is assigned on account on the host computer. This includes assignment of an (e-mail name, or address, such as [email protected] and allocation of some disk space on the host computer. For fusing the Internet facility, the user dials up the host computer by using his/her computer and the modem and then logs in to the host computer by using his/her email name and password. The user then uses the host computer's Internet software to access the Internet. The actual level of user's Internet access is dependent on the Internet software running on the host (different hosts may offer different packages of services - e-mail, FTP, telnet, Usenet news, gopher, WWW, etc.). Moreover, the user is also required to learn a little about the host computer': operating system to take advantage of his/her connection. So if the host is a Unix computer, the user will have to learn a few Unix commands to access the Internet. Essentially, the
user's computer acts as a dumb terminal to simply access the host and its Internet software. With this kind of connection, if a user visits an Internet site and downloads some information, the downloaded information gets stored in the disk space allocated to the user on the host computer. The user then needs to copy it to his/her personal computer by using the telecommunications software on his/her computer. That is, with a remote dial-up connection, anything a user uploads or downloads has to go first through the host computer. Hence the host computer acts 4s a transfer station. Fees for remote dial-up connections are normally based on hourly usage of the host computer by clients. SLIP/PPP Connection SLIP/PPP connection is an extension to the remote dial-up connection to allow client computers to have and run their own Internet software to access the Internet rather than depending on the host computer's software. With this type of connection, a user needs to know only about his/her own computer's operating system and Internet software and need not learn about the host computer's operating system and other software to take advantage of his/her connection to the Internet. This is made possible by using SLIP (Serial Line Internet Protocol) or PPP (Point-to-Point Protocol), which are software that allow a client computer to be connected via a modem to a he computer, and once connected, the client computer becomes an actual domain on the Internet. Hence all t software tools that the user of a client computer uses for interfacing with the Internet are the ones that are put 1 and run on the client computer. For a SLIP/PPP connection, a user requires a computer with TCP/IP (Transmission Control Protocol/Intent Protocol) communication software, a telephone line, a fast modem (28.8 Kilobits/sec or faster), and SLIP or PPP software. When a user wants to set up a SLIP/PPP connection, the Internet Service Provider (ISP) asks to specify which of the two software the user wishes to use and, depending on the user's choice, it provides one of the two software free of charge to the user. SLIP and PPP provide essentially the same functionality, but the gene opinion is that PPP is newer, better implemented, and likely to replace SLIP in the long run; however, SLIP more common today. If you have the option, go with PPP. The user is also responsible for acquiring a installing the Internet software tools that he/she wishes to use for accessing various Internet services. Like for remote dial-up connections, fees for SLIP/PPP connections are also normally based on hourly usage the host computer by clients.
Commercial Online Services At the time of writing this, this method of Internet connectivity is not available in India. It is used in some of the advanced countries like the US. In this method, a service organization offers a complete online environment access to some or all of the Internet's services by using phone service to connect a client's local call to the host computer of the service organization. The service organization offers local phone access to most areas of country. This means that if a client is traveling with a laptop, he/she can usually access his/her online account from different cities with only a local phone call. There are a handful of such service organizations in the US that provide commercial online services. The popular ones are CompuServe, America Online (AOL), GEnie, Delphi, eWorld, and Prodigy. A client has to register with one of the commercial online services to use the services provided by it. At the time of registration, the client is given an online account on the host computer of the commercial online services and he/she is given (usually on a floppy disk) all the software that is needed to access the service (including the telecommunications software). The client only needs to have a computer and a modem. The client logs in to his/her account by entering the phone number given to him/her by the commercial online services. Once logged in, the client can access the Internet through a graphical user interface. OTHER RELATED CONCEPTS Intranet Most organizations often require their employees to have easy and quick access to department, corporate, and customer information. To address this requirement, they need more dynamic ways to link people and information than conventional means of communication. Several organizations haye started using Intranets to fulfill this requirement. An Intranet is a network that connects the computers and networks within an organization by using the hardware and software that is used on the Internet. Typically, it uses the TCP/IP protocols and server and browser software used for the Internet. With an Intranet, the basic services of the Internet (electronic mail, file transfer protocol, telnet and usenet news) becomes available to the employees and business partners of the organization. This allows the employees and business partners of the organization to communicate effectively and quickly by using electronic mail facility, and to share information, software and other computing resources by using the other facilities.
The various departments of the organization can put the information on their web pages on to the Intranet, and all interested parties can access the information, read it, and download it without wasting the paper to copy it, or the time to make and circulate the copies. Similarly, employees of the various offices of the organization located across the world can form a newsgroup to brainstorm or exchange ideas collectively on topics of common interest. An Intranet being confined to the computers and networks of an organization, the information on the Intranet is confined to the employees and business partners of the organization. This is a great security advantage of Intranets. Furthermore, since an Intranet uses the same software as the Internet, which most computer users are already familiar with, the users of the Intranet do not require any additional training. Owing to all these features, the usage of Intranets is growing rapidly to more and more organizations. Proxy Server An organization's network interconnects all the computers of the organization. If all these computers are directly connected to the Internet, maintaining security becomes difficult for the organization because the Internet is an open environment. There may be virus attacks on the organization's computing resources and other security problems through the Internet. To handle this problem, organizations often use a proxy server, which is a computer system that is specially configured to allow other computers of an organization to interact with the Internet through it. When only a single proxy server interacts with the Internet, security can be easily maintained because the single server can be kept more secure than hundreds of individual computers on a network. With proxy server, the Internet is accessed from within the organization in the following manner. When someone from the organization specifies the URL of a server on the Internet using the browser on his/her computer, the user's request is first sent to the proxy server. The proxy server then contacts the server on the Internet and then the proxy server sends the requested information from the Internet server to the user's computer. All this is transparent to the user. That is, the user gets a feeling that he/she is directly accessing the Internet server specified in the URL. By acting as an in-between system, the proxy server can maintain security as well as log all traffic between the Internet and the organization's network of computers. A proxy server is typically used in an organization to provide one or more of the following services: 1. To serve as a firewall. A firewall controls the flow of incoming and outgoing messages from within an organization for providing better security to the
organization. For example, the firewall software running on the proxy server can be configured for packet filtering to ensure that only certain packets are permitted to flow from the inside to the outside world. Similarly, the firewall software can be configured to allow telnet to a computer within the organization only from those computers of the outside world whose IP addresses have been pre-registered in the proxy server. Many firewalls also permit establishment of port filters on a per host or subnet basis, allowing the network administrator to control access on a fine grained basis than just the whole organization. 2. To log the Internet traffic. Proxy servers can be configured to log every IP address, date and time of access, URL, number of bytes downloaded, and so on. This information can be used to analyze any attacks against the organization's systems. It can also be used to analyze who spends how much time on the Internet and what types of information does he/she accesses. 3. To scan viruses entering through the Internet. Proxy servers can be configured to scan all incoming information for potential virus attacks through the Internet. This protects the organization's computers from virus attacks because many computer viruses penetrate an organization's systems via the Internet. 4. To cache recently accessed web pages. Proxy servers can also be used to speed up performance of some Internet services by caching recently accessed web pages on its local hard disk. If someone within the organization makes a request for a web page which is already existing in the cache, then the proxy server sends the web page to the requesting user's computer directly from its cache rather than obtaining the web page from the specified Internet server and then forwarding it to the user's computer. This leads to a fairly high speed of Internet access in many situations. Internet Telephony Internet telephony refers to the use of the public Internet for voice transmission. That is, it enables the subscriber of an Internet service to use his/her computer to make long-distance calls. For this, the subscriber's computer must have a full duplex sound card and high-quality microphone. Both computer to computer and computer to telephone calls over the Internet are made possible with this facility. In addition to the hardware mentioned above, Internet telephony software is needed for this facility. Some of the commonly known Internet telephony software packages are Net2phone, Dial pad, Hot telephone and Phone Free. Some of these packages enable computer to computer voice communication, while others enable computer to telephone communication. Internet telephony makes use of the Internet Protocol (IP) that computers on the Internet use to communicate with each other. The parties involved in a typical setup for a computer to voice call include 1. The subscriber (the users),
2. 3.
The Voice Over IP (VOIP) service provider, and The user's Internet Service Provider (ISP)
That is, the user needs to first get registered with both the ISP and VOIP service provider. Now a call using Internet telephony takes place in the following manner: 1. The user makes a call using his/her computer. 2. He/she first gets connected and authenticated by the ISP, which provides the gateway to access the Internet. 3. The user is then connected to the VOIP service provider's gateway. This gateway has the Internet telephony software, which converts the packets into voice form and connects to the telephone network of the destination partly. The called party's telephone will then ring. 4. Once the call is established, the two gateways will constantly communicate and ensure that the packets are forwarded with the least loss. Microbrowser With the growing success of mobile communication technology and proliferation of mobile devices, such as mobile phones, Personal Digital Assistants (PDAs) and two-way pagers, there is a growing demand for Internet-enabled hand-held devices. To meet this demand, several vendors are now producing Internet-enabled hand-held devices. These devices allow users to access e-mail and web pages on their mobile devices. Microbrowsers are browsers meant for use with Internetenabled mobile devices. They are designed to work with small, resource-restricted hand-held devices by, for example, eliminating or reducing the functionality found in traditional desktop browsers. Because hand-held devices have limited processing and memory resources, microbrowsers are designed to be small so that their size is only a fraction of the size of their desktop counterparts. Thus, while desktop browsers occupy around 20 Mbytes, most microbrowsers occupy only a few hundred Kbytes. Their size ranges from 64 Kbytes to 2 Mbytes. Many vendors achieve these small sizes by providing simplified interfaces and by eliminating much of the desktop browser's multimedia functionality such as streaming-audio and -video support. Microbrowsers take one of the following two general approaches to accessing web data: 1. Direct Access Approach. The microbrowsers based on this approach use an HTML-based language, such as compact HTML and XHTML. Compact HTML, an HTML subset, works with small devices' resource limitations by not supporting, for example, JPEG images, tables, multiple fonts, or more than two colors. XHTML is basically HTML reformulated as an XML application. XHTML
lets a small device access just a subset of a web page for easier downloading and display. As shown in Figure 18.5(a), in case of microbrowsers that use this approach, clients get data directly from a web server without any intervention. However, this approach is limited by the server's ability to send information in HTML. A few microbrowsers that use this approach are Microsoft's Mobile Explorer, the Opera browser, and Pumatech's Browse-it. 2. WAP Gateway Approach. The microbrowers based on this approach use WAP's Wireless Markup Language (WML). As shown in Figure 18.5(b), in this approach a content-adaptation server, known as WAP gateway, is used. The data downloaded from a web server first goes to the WAP gateway. The WAP gateway translates the web content written in HTML or other languages into WAP' WML. Thus the WAP gateway, which acts as a proxy, translates and formats the web content i WML for the target hand-held device. A few microbrowsers that use this approach are Neomar Microbrowser and the Openwave Mobil Browser. Next Generation Internet Protocol (IPV6) The Internet connects together several hundred thousands of networks, and several tens of millions of computers. Several millions of users access the Internet from over more than hundred countries. The size of the Internet over the last few years has grown and continues to grow exponentially. Presently, the Internet works on IPV4 (Internet Protocol Version 4), which is popularly known as IP.IPV4 has been around since 1978. By the middle of 1990s, it was recognized that there are several limitations in IPV4 i a new version of Internet Protocol (IP) needs to be worked out. Some of the important limitations identified IPV4 are: 1. The 32-bit address space of IPV4 is not sufficient to cope up with the growing size of the Internet and its user community. That is, it would not permit enough Internet addresses for the rapidly growing number of users. What was sufficient for a few million users or a few thousands of networks, will of longer be sufficient to support a world with tens of billions of nodes and hundreds of million of networks. 2. When IPV4 was defined, multimedia applications were not so popular. Hence IPV4 lacks feature support multimedia applications that need real-time services or time-sensitive services.
3. IPV4 has no support for security at IP layer. With several Internet applications dealing with critical data, security has become an important issue. 4. IPV4 has limited mobility support, which has become a need for the users dealing with mobile applications. To address these limitations of IPV4, the Internet Engineering Task Force (IETF) developed and proposed a new version of the Internet Protocol. This version is popularly known as IPV6 (Internet Protocol Version 6) or Next Generation IP. The key features of IPV6 are: 1. It uses 128-bit address space instead of 32-bit in IPV4, which is believed to ensure that the world will not run out of IP addresses. With this large address space, it is expected that the Internet under IPV6 can support 1015 (quadrillion) hosts and 1012 (trillion) networks (the Internet under IPV4 can support maximum 232 hosts). Therefore the IPV6 address space is approximately 64 x 109 times more than that of IPV4. 2. It uses flow labeling and priority concepts to support real-time services. Senders can label packets that belong to high-priority traffic. By assigning higher priority to the packets belonging to real-time applications, the necessity of time sensitiveness is restored. Packets that are time insensitive are assigned low priority and serviced by the best effort approach. Senders can also request special handling of packets by the routers by assigning flow labels to them, such as a nondefault quality of service or real-time service. With this feature, video and audio data may be treated as flows whereas traditional data, such as file transfer and email, may not be treated as flows. 3. IPV6 with its ancillary security protocol provides a better communication tool for secure business transaction over the Internet. With a longer address field, IPV6 permits extensions that would natively support authentication, data integrity, and data confidentiality, which would address a variety of critical security concerns. 4. The enhanced routing capability of IPV6 includes support for mobile hosts. This feature allows a node to move from one link to another without changing the address. Movement can even be heterogeneous, that is, a node can move from an Ethernet link to a cellular packet network. 5. Besides unicast and multicast, IPV6 has the provision of anycast addressing. Anycast addressing allows a packet addressed to an anycast address to be delivered to any one of a group of hosts. 6. IPV6 also permits auto-configuration. IP-enabled devices can thus gain full access to the Internet without requiring time-consuming manual configuration. 7. IPV6 uses improved packet header format by dropping the least used options from IPV4 packet header format. IT also has improved support for extensions that brings greater flexibility in operations. All these features permit more efficient packet handling.
Points to Remember 1. The Internet is a huge network of computers that links many different types of computers all over the world. It is a network of networks that share a common mechanism for addressing (identifying) computers and a common set of communication protocols for communications between two computers on the network. 2. The four basic services provided by the Internet to its users are electronic mail, file transfer protocol, telnet, and usenet news. 3. The electronic mail service (known as e-mail in short) allows an Internet user to send a mail (message) to another Internet user in any part of the world in a nearreal-time manner. 4. The File Transfer Protocol service (known as FTP in short) allows an Internet user to move a file from one computer to another on the Internet. 5. The telnet service allows an Internet user to log in to another computer somewhere on the Internet. 6. The usenet service allows a group of Internet users to exchange their views/ideas/information on some common topic that is of interest to all the members belonging to the group. 7. There are several user-friendly tools that allow users to successfully navigate the Internet to find useful information. A few of the most popular of these tools are Gopher, Archie, and the World Wide Web (WWW). 8. The WWW uses a concept called hypertext. Hypertext documents on the Internet are known as Web Pages. Web Pages are created by using a special language called HyperText Markup Language (HTML in short), which is a powerful language for linking documents for easier electronic access and manipulation of documents. 9. To be used as a web client, a computer needs to be loaded with a special software tool that is known as a browser. Browsers normally provide several navigation facilities to help users save time when they are doing Internet surfing. 10. The three common types of browsers are line browsers, graphical browsers, and Java-enabled browsers. 11. An Internet search engine is an application available on the WWW that helps users locate the web sites containing useful information and references to such information. 12. Some of the important current strategic uses of the Internet are in on-line communication, software sharing, exchange of views on topics of common interest, posting of information of general interest, product promotion, feedback about products, customer support service, on-line journals and magazines, on-line shopping, and world-wide video conferencing. 13. The four basic ways to get connected to the Internet are: direct/leased-line connection, remote-dial-up connection, SLIP/PPP connection, and commercial online service.
14. The next generation Internet Protocol (popularly known as IPV6) has been developed to address the limitations of the current Internet Protocol (IPV4). The key features of IPV6 are large address space to ensure that the world will not run out of IP addresses; flow labeling and priority concepts to support real-time services; better security features for secure business transactions; enhanced routing capability to support mobile hosts; and provision for anycast addressing autoconfiguration and improved packet header format. 15. An Intranet is a network that connects the computers and networks within an organization by using the hardware and software that is used on the Internet. Typically, it uses the TCP/IP protocols and server and browsers software used for the Internet. 16. For security reasons, organizations often use a. proxy server, which is a computer system that is specially configured to allow other computers of an organization to interact with the Internet through it. When only a single proxy server interacts with the Internet, security can be easily maintained because the single server can be kept more secure than hundreds of individual computers on a network. 17. Internet telephony refers to the use of the public Internet for voice transmission. It enables the subscriber of an Internet service to use his/her computer to make long-distance calls. 18. Microbrowsers are browsers meant for use with Internet-enabled mobile devices. They are designed to work with small, resource-restricted hand-held* devices by, for example, eliminating or reducing the functionality found in traditional desk-top browsers. Questions 1. What is the Internet? How did it evolve? 2. List out three main characteristic features of the Internet that may be used to define it. 3. Name some of the basic services provided by the Internet. Explain how each of these services helps the Internet users. 4. In what manner is e-mail service similar to postal mail service? In what manner are the two different from each other? 5. List out some of the advantages and disadvantages of e-mail service against fax service. 6. List out some of the advantages and disadvantages of e-mail service against telephone service. 7. What is an electronic mail (e-mail)? Why is it preferred by many to paper mail, telephone and fax services? 8. What is the file transfer protocol (FTP)? List out the steps involved in downloading/uploading a file by using the FTP service. ' —
9. Explain how is it ensured that only authorized users can access resources of a remote computer in case of FTP and Telnet services. 10. Explain the difference between "downloading" and "uploading" of information. 11. List out some of the common uses of the telnet service offered by the Internet. 12. What is an anonymous FTP site? Why is it used? 13. What is a newsgroup? How is it useful? 14. Differentiate between moderated and nonmoderated newsgroups. 15. Answer the following questions: (a) How is a news group created? (b) How does a new member subscribe to a news group? (c) How does an old member unsubscribe to a news group? (d) How is a new message posted to a news group? (e) How is a message read from a news group? 16. What is meant by Internet surfing? Name some of the commonly used surfing tools. 17. What is hypertext? How is it useful? 18. Define the following terms with respect to the Internet: (a) HTML (d) Web client (b) HTTP (e) Web browser (c) Web server 19. What is a WWW browser? What types of navigation facilities are typically supported by modern browsers to help users save time while Internet surfing? 20. Differentiate between a line browser and a graphical browser. 21. What is a Java-enabled browser? In what manner is it better than conventional graphical browsers? 22. Explain how the following features are supported by a Java-enabled browser: (a) Dynamic content, (b) Dynamic data types, and (c) Dynamic protocols. 23. What is an Internet search engine? What are the three major elements of an Internet search engine? Explaii the roles of these elements. 24. What are the major categories in which Internet search engines are normally categorized? Describe thi features of each of these categories of Internet search engines. 25. Name three popular Internet search engines. Describe some of their key features. 26. Describe some of the typical uses of the Internet. 27. Write short notes on: (a) Electronic mail (c) Telnet service (b) File transfer protocol (d) Usenet news 28. Write short notes on:
(a) Internet search tools (b) WWW browsers (c) Uses of the Internet 29. Write short notes on: (a) Netiquette (b) Internet surfing (c) Gopher (d) Ways to get connected to the Internet (e) Internet search engines (d) Archie (e) WWW 30. List out three different ways in which a user can get his/her computer connected to the Internet. List out t relative advantages and limitations of the three methods. 31. What is leased line connectivity? For what type of users is it normally used? 32. What is an Internet Service Provider (ISP)? How does it provide Internet connectivity to the users? 33. Differentiate between the method of remote dial-up connection and SLIP/PPP connection of getting connected to the Internet. List out their relative advantages and limitations. 34. Write short notes on: (a) Leased-line connectivity (c) SLIP/PPP connectivity (b) ISP (d) Commercial online services 35. What is IPV6? Why has it been developed and proposed? 36. List out some of the main limitations of the current Internet Protocol. Explain how the newly proposed Internet Protocol takes care of these limitations. 37. List out the main features of IPV6. 38. What is an Intranet? How is it different from the Internet? How is it similar to the Internet. 39. Explain three different ways in which an Intranet may be useful to an organization. 40. What is a proxy server? Why is it used? Explain how the Internet is accessed from within an organization that uses a proxy server. 41. List out some of the typical uses of a proxy server. 42. What is a firewall? Explain how can it help in providing better security to an organization. 43. What is Internet telephony? Explain how a call takes place by using Internet telephony. 44. What is a microbrowser? How is it different from a traditional desktop browser? What are the two general approaches used by microbrowsers for accessing web data? (e) Next generation Internet Protocol (IPV6) 45 Write short notes on:
. (a) Intranet (b) Proxy server (c) Internet telephony 46 Give the full form of: . (a) FTP (b) FAQ (c) WWW (d) HTML (e) SGML (f) HTTP (g) ISP
(d) Mien (e) Next
(h) URL (i) SLIP (j) PPP (k) VOIP (1) IPV6 (m)IETF
Chapter 19 Multimedia You saw in Chapter 1 that modern computers are not only used as a fast calculating machine, but also as a device that operates upon data. You also saw that data comes in various shapes and sizes depending on the type of computer application. - In fact, the increasing popularity and usage of computers over the years is mainly due to their capability to handle different types of data. This capability of computers makes them suitable for a very wide range of applications and users. Hence, this chapter will introduce you to the different types of data that modern computer systems can process and the issues involved in processing these data types. This chapter starts with an introduction to multimedia and multimedia computer systems. It then describes the commonly used media in computer systems along with the issues in handling each one of them. Some key multimedia applications are then described to allow the readers to appreciate the various uses of multimedia. Finally, the chapter also covers some advanced topics on multimedia like data compression techniques and multimedia synchronization. Beginners may skip these advanced topics. WHAT IS MULTIMEDIA? A "media" is something, which can be used for presentation of information. There are two basic ways to present some information: 1. Unimedia presentation. In this case a single media is used to present information. For example, stereo system is a unimedia device because it presents information by using only sound media. Similarly, a book having only text (no diagrams) is also a unimedia device because it presen information by using only text media. 2. Multimedia presentation. In this case more than one media is used to present information. F example, a TV system is a multimedia device because it presents information by using sound and video media. Similarly, a book having both text and diagrams is also a multimedia device because it presents information by using text, graphics and images. Obviously, multimedia presentation of any information greatly enhances the comprehension capability of the user of the information because it involves the use of more of our senses. For example, students in a class can understand a lecture better if the lecturer draws illustrative diagrams on the board while delivering his/her lecture (multimedia presentation) instead of using only voice explanation (unimedia presentation).
In case of computer systems, the commonly used media for the purpose of storage, access and transmission of information are: 1. Text (alphanumeric characters) 2. Graphics (line drawings and images) 3. Animation (moving images) 4. Audio (sound) 5. Video (Videographed real-life events) Hence, in the computer world, multimedia refers to the use of more than one of the above listed media for information presentation to the computer users - such as text and sound. WHAT IS A MULTIMEDIA COMPUTER SYSTEM? A multimedia computer system is a computer system that has the capability to integrate two or more types of media (text, graphics, image, audio, and video) for the purpose of generation, storage, representation, manipulation and access of multimedia information. Advances in computer hardware and software have led to the development of inexpensive input, output, storage and processing devices for multimedia computer systems. Technically, multimedia on computer systems is not as new as many people think. For example, computer systems capable of displaying graphs (such as sales graphs), charts (such as bar charts, pie charts), maps (such as contour maps) and engineering drawings (such as mechanical drawings, building plans, circuit diagrams, perspective views) along with text information have been in use for several years. What is more exciting and relatively new about multimedia computer systems is their enhanced capability to handle additional media like sound and video for better information presentation. In general, the data size for multimedia information is much larger than plain text information because representation of graphics, animation, audio or video media in digital form requires much larger number of bits than that required for representation of plain text. Because of this, multimedia computer systems require: 1. Faster CPU (for more quickly processing large amount of data), 2. Larger storage devices (for storing large data files), 3. Larger main memory (for running programs with large data size), 4. Good graphics terminals (for displaying graphics, animation and video), and 5. I/O devices required to play any audio associated with a multimedia application program. Note that logically speaking, all multimedia computer systems need not have all the features listed above. For example, a computer system that does not have the capability to handle audio and video media can still be called a multimedia computer system because it can still handle multiple media (text and graphics).
However, a full-fledged multimedia computer system must be capable of handling all types of media discussed above and hen must have all the features listed above. MULTIMEDIA COMPONENTS Text Alphanumeric characters are used to present information in text form. Computers are widely used for text processing. Hardware Requirements for Text Text processing using computers generally involves the following hardware devices (see Chapter 9 for a description of some of these I/O devices): 1. Keyboards are most commonly used to input text data. 2. OCRs (Optical Character Recognizers) are used to codify printed text for direct input of printed text to computers. 3. Computer screens are used to display text information. 4. Printers are most commonly used to output text in hard copy form. Software Requirements for Text In addition to the above listed devices, the following text processing capabilities are highly desirable ii multimedia computer system for better presentation and use of text information: 1. Text editing. Text editors and word processing packages are used to generate, edit and props layout a text document. 2. Text style. Presentation of text information can be made more effective by using text of various sizes, fonts and styles (bold, italics, shadow, etc.). For example, newspapers use a good combination of all these styles to better highlight more important news than less important ones. Similarly, book like the one you are now reading, a combination of various text styles are used to put more emphasis on certain words or phrases than the remaining text. 3. Text searching. Usage of text information can be greatly enhanced by text searching feature. This feature allows the user to enter a word or phrase and quickly finds and displays that part of the where the word or phrase appears in the text information with the word or phrase highlighted. It is equivalent to searching for a word in a book's index, and then going to the indicated page or pages for more information. However, this process is much faster in a computerized text information. Almost all word processing packages have this feature. Several software having this feature are also available on the Internet that help in searching desired information from the Internet.
4. Hypertext. Both presentation and use of text information can be greatly enhanced by using hypertext feature. This feature provides great flexibility of structuring and accessing computer-based text information. It generalizes the concepts of "footnote" and "cross reference" of traditional information (or document) organization and allows users to obtain information just by clicking on an anchor (a word or phrase linked to another document) within a document. An anchor is often called a "hotword". This means that if the user wants more information about a particular hotword, he/she can select that hotword (usually with a mouse pointer) and open a window with additional text explaining that word or phrase. Figure 19.1 shows a typical structure of hypertext documents in which information units are stored in a network of nodes interconnected by links. This type of information structuring allows nonsequential navigation within the information space according to individual needs. The author of a hypertext document determines the actual structure and navigation options (by defining hotwords in the document and linking them to additional text information), and the readers decide the order of navigation according to their individual interests while reading the document online. Generally, hotwords are distinguished from other text displayed on the screen. It may be of different color or font, or the mouse pointer may change into a special symbol (such as a hand) when moved over the hotword. This makes it easier for the user to identify which words have associated links with more information. Otherwise, the user can waste a lot of time clicking on words that have no other associated information. 5. Text importing and exporting. The task of creating a text document can often be greatly simplified if the document preparation software has text importing feature. This is because some of the text you want to incorporate in your document may already exist as a file created by a word processor or a database file. The file (partly or fully) can be simply imported into the new document at the desired location instead of retyping the entire text again. If the imported file is large, it will result in enormous saving of time and effort required to prepare the new document. Similarly, text exporting feature is very useful in allowing other packages to use an existing document (partly or fully). Graphics Computer graphics deals with the generation, representation, manipulation and display of pictures with the aid of a computer. Graphics is an important component of multimedia because a picture is a powerful way to illustrate information. For example, in a multimedia application meant for educating small children, displaying the photograph of a horse along with its text description makes the illustration much more effective. Without the photograph, it will be very difficult
to explain using only text how a horse looks like. Similarly, displaying the photograph of an employee, along with his/her name, address, sex, age, etc. from a company's database of employees, makes the employee information much more effective. Graphics Types The pictures used in computer graphics can be broadly classified into two types: 1. Line drawings. These are drawings and illustrations in the form of 2D and 3D pictures created from mathematical representation of simple objects like lines, circles, arcs, etc. Simple object types are used to create complex objects. For example, the picture of a chair can be drawn using lines and arcs. Numerical results obtained from computations can also be transformed to graphics forms like bar graphs or pie charts using simple objects. The area of computer graphics that deals with this type of pictures is known as generative graphics. Generative graphics is extensively used in creating illustrative diagrams. For example, most of the figures in this book have been created by using this technique. Another very important application of generative graphics is CAD (Computer Aided Design) and CAM (Computer Aided Manufacturing). For example, nowadays CAD packages are extensively used in designing models for aircraft, ship, and building structures, layout of printed circuit boards, and chemical plant pipelines, etc. Based on these drawings, manufacturing information such as parts list, subassembly list, etc., are automatically generated and used by CAM packages. The greatest advantage of using CAD and CAM packages for such applications is that design ideas may be quickly checked and graphically conceptualized interactively, and design changes may be quickly modified, immediately creating accurate assembly drawings and associated parts and subassembly lists. 2. Images. These are pictures and photographs composed of a collection of pixels (short form of "picture element," which is a unit of measurement across a computer screen). As shown in Figure 19.2, the pixels of an image are arranged as a two-dimensional matrix. This two-dimensional (or spatial) representation is called the image resolution. Since the storage representation of images is a direct translation of the image, pixel by pixel, no concept of a line, arc, or circle exists. As shown in the figure, each pixel consists of three components: the red (R), the green (G), and the blue (B). On a display screen, each component of a pixel corresponds to a phosphor. A phosphor glows when excited by an electron gun. Various combinations of different RGB intensities produce different colors. The number of bits used to represent a pixel is called the color depth, which decides the actual number of colors available to represent a pixel. Color depth in turn is
determined by the size of the video buffer in the display circuitry. For example, a screen with a 640-by-480 resolution and with 24-b color (i.e., 8 bits each for the R, G, and B components) displays 640 pixels horizontally and 480 vertically, and each pixel can be represented by selecting from more than 224 (^ 16.7 million) different colors. The resolution and color depth determine the presentation quality and the size of image storage. The more pixels and the more colors, the better the quality and the larger the volume. For example, a 640-by-480 image with 24-b color requires 640 x 480 x 24b = 7.4 Mb of storage. To reduce the storage requirement, images are usually compressed before being stored or transmitted. Image compression techniques are discussed later in this chapter. The area of computer graphics that deals with this type of pictures is known as cognitive graphics. Cognitive graphics along with image processing techniques (like image compression and decompression, cleaning of noisy or blurred images, image editing, etc.) is extensively used in applications that deal with recognition and classification of pictures. For example, an image database containing images of peoples' fingerprints is commonly used nowadays in criminal investigation. Another similar application is classification of images of ECG patterns, X-rays of different parts of the body, etc. to perform medical diagnosis of patients by using computers. Hardware Requirements for Graphics Computer graphics generally involves the following hardware devices (see Chapter 9 for a description of some of these I/O devices): 1. A locating device (such as a mouse, a joystick, or a stylus) attached to a video display terminal along with a drawing software is often used to directly draw pictures on computer screen from scratch for generative graphics applications. 2. A flatbed or rectangular-coordinate digitizer is also used as an input device for generative graphics applications for inputting existing line drawings (such as a map) to a computer. 3. Scanners (optical scanners, image-scan digitizers, etc.) are most commonly used as input devices for capturing photographs and drawings as digital images for cognitive graphics applications. 4. Digital images are also captured directly by a digital camera or a frame capture hardware such as a video capture board interfaced to some standard video source like video camera or video cassette recorder (VCR). 5. Computer screens with graphics display capability are used to display graphics. Standard VGA (Video Graphics Array) graphics can display 640 pixels across and 480 pixels down (referred to as having a resolution of 640 x 480) on the monitor with 16 colors (or gray scales, if the monitor is a monochrome monitor). Although there are higher resolutions available (for example, Super VGA has a
resolution of 800 x 600), Standard VGA can display very good images and comes with most new computers. Standard VGA requires a VGA board and a monitor capable of displaying VGA graphics. A Standard VGA board comes with 256K of memory attached to the board. This limits 640 x 480 VGA displays to 16 colors. Both the resolution and the number of colors of display can be increased by increasing the memory on the graphics board. For example, if this memory is increased to 512K, a resolution of 640 x 480 with 256 colors is possible. Similarly, if this memory is increased to 1 MB, resolutions of 800 x 600 or 1024 x 768 with 256 colors are possible. These graphics are called extended graphics. Higher resolutions with more colors make image displays noticeably better. But the computer system must have a monitor capable of displaying such high-resolution images. 6. Laser printers are*most commonly used to output graphics in hard copy form. 7. Plotters are also used to output graphics in hard copy form. Software Requirements for Graphics In addition to the above listed devices, the following graphics processing capabilities are highly desirable ii multimedia computer system for better presentation and use of graphics information: 1. Painting or drawing software. This software allows the user to create graphics from scratch by using a mouse and various simple objects such as lines, circles, and polygons with various supporting colors. It also allows the user to modify and enhance graphic images produced using a scanner or screen capture software. All these features make this software very useful for creating desired graphics in a multimedia application. 2. Screen capture software. Often we need to incorporate images from computer screen displays in some document. Screen capture software packages are used in such cases to capture the display of a computer screen as a graphic image. These packages allow the users to save the graphic images in a variety of graphic formats and are also useful in converting a graphic image from one storage format to another because a multimedia software might have been designed to accept image data only in a particular format. 3. Clip art. Clip art is a library of commonly used graphic images or objects, such as a personal computer, printer, aeroplane, telephone, flower, building, etc. These images can be directly imported from the library and used in a multimedia application. This saves enormous time and effort that might otherwise be required to create or search and capture a similar image object. This is also helpful if one does not have a scanner or some other means of inputting graphics into the computer. A clip art library often provides the facility to add new objects or delete an existing object from the library. 4. Graphics importing. The task of creating a multimedia application incorporating graphics can offer be greatly simplified if the application software
can import graphic images in some standard format. Common graphic formats include .BMP, .GIF, and .PCX. 5. Software support for high resolution. If a multimedia application needs to display very high quality images, it is important to have not only the necessary hardware but also software support for high resolution graphics. Some software cannot display a resolution greater than 640 x 480 pixels. Others cannot display more than 16 colors. Most software, however, support Standard VGA graphic images. Hence, extended graphics hardware feature must also have the right software for displaying higher resolution graphics with 256 colors. Animation Computer animation deals with the generation, sequencing, and display (at a specified rate) of a set of images (called frames) to create an effect of visual change or motion, similar to a movie film (video). Animation is an important component of multimedia because just as a picture is a powerful way to illustrate information, a small animation clip is even more powerful and is very useful for illustrating concepts that involve movement. For example, in the multimedia application meant for educating small children (our previous example), an animation clip will be very effective in illustrating the difference in movement of a horse during a gallop versus a walk. Without animation, it will be very difficult to explain such concepts using only still images and/or text. Similarly, in a multimedia application that contains training materials for teaching swimming, the proper way to swing hands for swimming can be more effectively illustrated with an animation as compared to images and/or text description. Animation is often confused with video (described later in this chapter) may be because both deal with the display of a sequence of images (frames) to generate an effect of motion. However, animation is commonly used in those instances where videography is not possible or animation can better illustrate the concept than video. For example, collision between two stars or movement of planets around sun cannot be videographed and can be best illustrated with the help of an animation clip. Similarly, flow of blood in our arteries and veins and pumping of blood by our heart can be best illustrated with the help of an animation clip. Animation is also very commonly used in advertisements, cartoon films, video games, and virtual reality applications. Virtual reality is a relatively new computer technology using which the user can put on a pair of goggles and a glove and tour a three-dimensional world that exists only in the computer, but appears realistic to the user. Animation deals with displaying a sequence of images at a reasonable speed to create an impression of movement. For a jerk-free full motion animation, 25 to 30 frames have to be displayed per second. We saw in the previous section that a 640-
by-480 image with 24-b color requires 7.4 Mbits of storage. Therefore, a one second of animation will require 7.4 x 25 = 185 Mbits =; 23 Mbytes (185 -=- 8) of storage. In general, an animation clip lasts for several seconds. Hence animation files require much more storage space than graphic files involving a single image. Moreover, to play an animation clip stored in a file on a secondary storage device like hard disk, it should be possible to read the data at the required rate and transfer it through the bus to the frame buffer to display it on the monitor. Hardware Requirements for Animation Computer animation generally involves the following hardware devices: 1. Image generation tools and devices such as scanners, digital camera, and video capture board interfaced to some standard video source like video camera or video cassette recorder (VCR) are used to generate images to be used in animation. 2. Computer monitors with image display capability (described in the previous section) is the minimum requirement for outputting (displaying) animation. In addition, multimedia computer systems capable of handling animation also require a graphics accelerator board that not only controls the resolution and number of colors of display, but also speeds up the refresh rate. Refresh rate is the time taken to paint the computer screen with a new image frame replacing the previous image frame. Increased speed due to the use of graphics accelerator board smoothes out the movement (avoids flickering of animated sequences) because subsequent frames that make up the animation can be displayed more quickly. Software Requirements for Animation In addition to the above listed tools and devices, the following capabilities are highly desirable in a multimedia computer system with animation facility: 1. Animation creation software. This software allows the user to create animation sequences from scratch by using a mouse and various simple objects such as lines, circles and polygons with various supporting colors. For example, an animation sequence to illustrate bouncing of a ball can be created as follows: - select a circle object and fill it with red color; - then start the recorder of the animation creation software; - then select the colored circle and drag it around the screen using a mouse, recording several positions for it on the screen (each time you release the mouse button, you record the object's position on the screen); - finally stop the recorder and use save command to save the animation. Now to test the recorded animation, start the player of the animation creation software and click on the colored ball. The ball will bounce across the screen
according to the positions recorded during the recording phase. The same bouncing sequence will repeat every time you click on the ball. 2. Screen capture software. Screen capture software packages are used to capture the displays of a computer screen as graphic images, which can be used in animation. 3. Animation clips. This is a library of animation clips from which one can select and directly import an animation clip and use it in a multimedia application. This saves enormous time and effort that might otherwise be required to create a similar animation clip. This is also useful if one does not have an animation creation software. An animation clips library often provides the facility to add a new animation clip or delete an existing animation clip from the library. 4. Animation file importing. The task of creating a multimedia application incorporating animation can often be greatly simplified if the application software can import animation files in some standard formats. Common animation file formats include .FLI and .FLC. 5. Software support for high resolution. If the animation sequences of a multimedia application are made up of very high quality images, it is important to have not only the necessary hardware but also software support for displaying high resolution images having many colors to maintain the high quality of animation display. 6. Recording and playback capabilities. A multimedia system with animation capability should allow the user to control the recording and display of an animation sequence. For example, it should provide the user with options to 'pause' and 'replay' the animation sequence. 7. Transition effects. Animation can be even more interesting if it is enhanced with transition effects such as fade-in and fade-out, layering, zooming, and rotation of objects. Audio Computer audio deals with synthesizing, recording, and playback of audio or sound with the aid of a computer. Audio is another very important component of multimedia because in many cases sound can substantially reinforce our understanding of information presented in other ways, and in some cases sound is the only way to provide the desired information. For example, in the multimedia application meant for educating small children (our previous example), voice explanation can be added to the animation clip describing what is being shown and pointing out what to look for that makes a horse's gallop different from a horse's walk. Similarly an animation clip can show the horse neighing and neighing sound can be added to effectively convey the type of sound horses make. Note that it is nearly impossible to provide an accurate description of a horse's neigh, a lion's roar, or a bird's chirp using text, and graphics is of no help in these cases. Hence sound is the only media that can accurately provide such information.
Analog and Digital Audio Audio information travels in natural media in the form of sound waves, which are analog in nature. In order for the computer to be able to understand audio information, sound waves must be converted from analog form to digital form. Transducer is a device capable of changing signals from one form to another. For example, microphone is an example of a transducer that converts sound waves into electrical signals. Conversely, loudspeaker is an example of a transducer that converts electrical signals into sound waves. Information can be represented as analog signals or digital signals. Analog signals exhibit a/ continuous tone of smooth fluctuations, while digital signals are composed of discrete values represented by numbers. The transformation between analog and digital signals is achieved by A/D (Analog-to-Digital) conversion and D/A (Digital-to-Analog) conversion. A/D conversion transforms an analog input (such as voltage) into a series of numeric representation by digitization. D/A conversion is the reverse process of A/D conversion. It transforms a sequence of discrete numbers back into continuous analog signal. Figure 19.3 shows the role of A/D and D/A converters in the acquisition and playback of analog signals by a computer system. A multimedia computer system capable of processing audio information requires a sound board (or sound card), which is equipped with A/D and D/A converters. In addition to performing the basic function of A/D and D/A conversions of audio signals, a sound board is generally also equipped with the following features: 1. It has a connector for speakers or headphones to allow users to listen to sound recordings. 2. It has a MIDI input port to input sound from an external MIDI device (such as an electronic keyboard or drums), a MIDI output port to output recorded sound to the MIDI device, and a MIDI synthesizer to translate the information transferred between the computer and the musical instrument. MIDI (Musical Instrument Digital Interface) is an international standard developed for digital music that determines the cabling, hardware, and communications protocol needed to connect a computer with electronic musical instruments and recording equipment. 3. Most sound boards also have a volume control knob to allow users to control sound volume according to their taste. Digital Audio Quality The quality of digital audio is characterized by the following three parameters: 1. Sampling rate. Analog-to-digital conversion transforms an analog signal into a series of numeric representation by digitization. Digitization is composed of sampling and quantization. Sampling take; snapshots of the analog waveform at
certain points in time, and the sampling rate determines how often the analog signal is digitized. According to Nyquist theorem, the sampling rate must be at least twice the highest frequency component of the analog waveform to reproduce the signal optionally. The higher the sampling rate, the truer the sound. Sampling rate is measured in kilohertz (KHz) am its typical values are 8, 11, 22, and 44.1 KHz. For example, the sampling rate of CD-quality audio is 44.1 KHz, and that of telephone-quality audio is 8 KHz. 2. Quantization. Quantization determines the digital measured value of the analog waveform at the sampling time. The larger the range of numbers used for quantization, the more gradations of the digitally sampled waveform can be represented, resulting in better quality audio. Quantization is measured in terms of number of bits and its typical values are 8, 12, and 16 bits. Figure 19.4 illustrates sampling and quantization of an analog signal. In Figure 19.4(a), the analog signal is sampled at periodic points in time. The sampled values are then represented by the closest quantization levels. The difference between the value of the closest quantization level and the real analog value is known as quantization error. In Figure 19.4(b), 2 bits are used to represent quantization, resulting in 4 possible levels, and in Figure 19.4(c), 3 bits are used, allowing 8 quantization levels. As the number of bits used to represent quantization increases, more quantization levels become possible resulting in reduced quantization error and increased audio quality. 3. Number of channels. It refers to the number of channels used for audio recording. Its typical values are one for mono systems, two for stereo systems, and 5.1 for Dolby surround-sound systems. The quality and the storage and bandwidth requirements of digital audio depend on the three parameters discussed I above. The higher the sampling rate, the more bits used to represent quantization, and the more channels, the better the quality of the digital audio, and the higher the storage and bandwidth requirements. For example, an 8 KHz sampling rate, 8 bits quantization, and mono recording will produce telephone-quality audio, and will require a bandwidth of 8000 x 8 x 1 = 64 Kb/s. On the other hand, a 44.1 KHz sampling rate, 16 bits quantization, and stereo recording will produce CD-quality audio but will require a bandwidth of 44100 x 16 x 2 = 1.4 Mb/s. In general, an audio clip lasts for several seconds. Hence audio files are very large requiring large storage space. For example, with reference to the example given above, a 20 seconds CD-quality audio clip will require 1.4 x 20 = 28 Mb = 3.5 Mbytes of storage. Again, digital audio compression or a compromise in quality can be applied to reduce the storage requirement. Hardware Requirements for Audio Computer audio generally involves the following hardware devices:
1. A sound board with all the features already described before. 2. Some type of input device (such as a microphone) is used for audio input to record a human voice or music or any type of sound in a computer. The A/D converter of the sound board takes care of digitizing the input sound. 3. Some type of output device (such as speakers or headphones) is used for audio output to listen to a recorded sound. The D/A converter of the sound board takes care of converting the sound from its digital to analog form. 4. MIDI devices may also be used as both input and output devices for audio data. 5. Synthesized sound can also be generated on a computer using keyboard (as interaction device) and sound sequencer software. 6. Sound editors are used to cut and paste sound sequences, to add special effects, and to create new sound sequences from existing sound sequences. 7. Audio mixers are used to combine multiple channels of sound with controls like synchronization points. Software Requirements for Audio In addition to the above listed tools and devices, the following capabilities are highly desirable in a multimedia computer system with audio facility: 1. Audio clips. This is a library of audio clips (pre-made sound effects, music, and narration) from which one can select and directly import an audio clip and use it in a multimedia application. This saves enormous time and effort that might otherwise be required to create a similar audio clip. This is also useful if one does not have the facility to create sound sequences. An audio clips library often provides the facility to add a new audio clip or delete an existing audio clip from the library. 2. Audio file importing. The task of creating a multimedia application incorporating audio can often be greatly simplified if the application software can import audio files in some standard formats. Common audio file formats include .WAV (Windows files), .MID (MIDI files), .VOC, and .INS. 3. Software support for high quality sound. If a multimedia application uses very high quality audio, to reproduce the sound effectively, it is important to have not only the necessary hardware but also software support for both recording and playback of high quality audio. Some software allows users to select the fidelity level by capturing sound using varying values of quantization and sampling rate. Some software also has a conversion utility that allows users to convert the sampling rate of audio files. 4. Recording and playback capabilities. A multimedia system with audio capability should allow the user to control the recording and playback of an audio sequence. For example, it should provide the user with options to "pause" and "replay" the sound sequence.
5. Text-to-speech conversion software. This software is used to convert written text into corresponding sound. 6. Speech-to-text conversion software. This software is used to convert speech into corresponding text. 7. Voice recognition software. This software is used to identify the speaker of a given voice by matching the voice with a set of voices stored in a database along with the details of the speaker of each voice. Video Like animation, computer video deals with the recording and display of a sequence of images at a reasonable speed to create an impression of movement. Each individual image of a sequence of images is called a frame. For a jerk-free full motion video, 25 to 30 frames have to be displayed per second. Like animation, video is also an important component of multimedia because it is very useful for illustrating concepts that involve movement. Although both animation and video deal with the display of a sequence of images (frames) to generate an effect of motion, but video typically deals with recording of a real-life event produced by a device such as a digital video recorder. Like animation data, video data also requires large storage space and needs to meet the real-time constraints on retrieval, delivery, and display. Again, digital video compression or a compromise in quality can be applied to reduce the storage requirement. Video information travels in natural media in the form of light waves, which are analog in nature. In order for the computer to be able to understand video information, light waves must be converted from analog form to digital form. Video camera is a transducer that is commonly used to convert light waves into electrical signals. Conversely, monitor is a transducer that is commonly used to convert electrical signals into light waves. Like audio, in case of video also the transformation between analog and digital signals is achieved by A/D and D/A conversions as shown in Figure 19.3. Analog-to-digital conversion of video also involves the sampling and quantization of analog signals as already explained before and shown in Figure 19.4. A multimedia computer system capable of processing video information requires a video board (or video card), which is equipped with A/D and D/A converters. Video Display Technologies As already mentioned before in this chapter (see Figure 19.2), an image is represented as a collection of pixels, which are arranged as a two-dimensional matrix. Each pixel is of the size of a small dot. Hence, an image can be thought of as made up of horizontal lines, each a row of individual dots (pixels). To paint an entire video image on a monitor screen, an electron gun illuminates every pixel as
it sweeps across the screen, one line at a time, from left to right and from top to bottom and then back on the top. This process, called raster scanning, has to be repeated several times every second (called refresh rate) to provide continuity of motion and to avoid flicker. A complete line travelled by the electron gun from left to right is called a scanline. In one pass of raster scanning, if the electron beams travel all the scanlines sequentially from top to bottom, it is called & progressive scan (see Figure 19.5(a)). On the other hand, if the electron beams travel only the odd scanlines in one pass and the even scanlines in the next pass, it is called an interlaced scan (see Figure 19.5(b)). For the interlaced scan, each pass is called & field, and two complete fields comprise a full display screen. When the electron beam has travelled the bottom-most scanline, it retraces back to the top. Similarly, when the electron beam reaches the end of a scanline, it retraces back to the beginning of the next scanline to be traveled. To prevent the retrace lines from being visible on the screen, blanking pulses are applied to the picture tube. The time taken to move the electron gun from the end of the bottom-most scanline to the beginning of the top-most scanline is called the vertical blanking interval, while the time taken to move it from the right end of one scanline to the left end of the next is called the horizontal blanking interval. The vertical blanking intervals are often used to embed other signals, such as synchronization information, in broadcast television (TV) systems. The two commonly used types of monitors for video display are TV monitors and computer monitors. TV monitors use interlaced scanning technique with a refresh rate of 60 fields per second or 30 frames per second (the actual refresh rate depends on the broadcasting standard used by the TV system - NTSC, PAL or SECAM described later in this chapter). On the other hand, computer monitors use progressive scanning technique with a refresh rate of 60 frames per second. The aspect ratio, which is the ratio of the horizontal to the vertical dimensions, is 4:3 for both TV and computer monitors. Note that as compared to a computer monitor, a TV monitor costs less but flickers often due to the lower refresh rate. Types of Video Signals We saw in Figure 19.2 that a pixel is composed of RGB components. Human vision is more sensitive to variation in the luminance (brightness) than in the chrominance (color difference). To take advantage of such deficiencies in the human eye, light can be separated into luminance and chrominance components instead of the RGB components. For example, the YUV model separates luminance from chrominance via the following linear transformation from the RGB system
Y = 0.30R + 0.59G + 0.11B U = 0.62R-0.52G-0.10B V = -0.15R-0.29G + 0.44B Where Y is the luminance component and (U, V) are the chrominance components. Based on these concepts of light representation, there are following two types of video signals (for both analog and digital videos) for storage and transmission: 1. Composite video signals. A composite video signal combines the luminance Y and chrominance C into a single stream to be stored and transmitted. Examples of analog videos using composite video signals include broadcast TV signals and VHS. Examples of digital videos using composite video signals include the D-2 and D-3 formats. 2. Component video signals. A component video signal separates the video into luminance Y and chrominance C streams. Component video signals are again of two types - the Y/C component and the RGB component. The Y/C component uses two separate channels to transport Y and C information, whereas the RGB component uses three channels to carry R, G, and B colors. Examples of analog videos using component video signals include Hi8 (it is of Y/C component type) and Beta SP (it is of RGB component type). Examples of digital videos using component video signals include the D-l, D-4, and D-5 formats. Out of the two, component video signals have better quality than composite video signals. Broadcast Television Standards The three main TV broadcasting standards are NTSC, PAL and SEC AM. All three are analog videos using composite video signals technology. They mainly differ in the way they encode the chrominance signal. They also differ in screen resolution and frame rate. Their characteristic features are briefly described below. 1. NTSC. NTSC stands for National Television Systems Committee. It is a 525 scanlines system that uses interlaced scanning technique with a refresh rate of 29.97 frames per second. In NTSC, the luminance signal is called the Y signal, and the chrominance signal is amplitude modulated onto the color subcarrier at two different phases - the / (in-phase) value at 0° and the Q (quadrature) value at 90°. NTSC is used in the US, Japan , and some Asian countries. 2. PAL. PAL stands for Phase Alternating Line. It is a 625 scanlines system that uses interlaced scanning technique with a refresh rate of 25 frames per second. In PAL, the composite video signal operates on the YUV model, where Y is the luminance component and (U, V) are the chrominance components. In PAL also, the chrominance signal is transmitted on a two-phase amplitude-modulated color subcarrier but it uses a more complex process called Phase Alternating Line,
which provides more bandwidth for chrominance modulation, yielding better color resolution. PAL is used in India and Western Europe (except France). 3. SECAM. The SEC AM system, developed in France, stands for Sequential Couleur Avec Memoire. It is a 625 scanlines system that uses interlaced scanning technique with a refresh rate of 25 frames per second. In SECAM, the chrominance signal is transmitted using a frequency-modulated color subcarrier, transmitting one of the color difference signals on every other scanline, and the other color difference signal on alternate scanlines. It also has better color resolution than NTSC. SECAM is used in France, Eastern Europe, Russia, and Middle East. A summary of the characteristic features of the three analog TV signal standards is given in Figure 19.6. Compared to NTSC, PAL and SECAM have better resolution but lower refresh rates and hence more flicker. Standards Features
NTSC
PAL
SECAM
Scanlines/frame Scanning technique Refresh rate (frames/second) Aspect ratio Transmission Modulation Used in
525 Interlaced 29.97
625 Interlaced 25
625 Interlaced 25
4:3 4:3 Analog Analog AM AM US, Japan, and India and Western some Asian Europe (except countries France) Figure 19.6. Features of the three broadcast TV standards.
4:3 Analog FM France, Eastern Europe, Russia, and Middle East
High Definition Television (HDTV) HDTV emerged out of the desire to build a video system that could deliver a more exciting viewing experience than that offered by current TV technologies. The major goals in designing such a system were to provide for improved picture resolution, improved color resolution, wide screen display, and large projection. Two major broadcasting systems that have evolved for commercial use out of this effort are 1. MUSE (Multiple SubNyquist Sampling Encoding) in Japan and 2. HD-MAC (High Definition Multiple Analog Components) in Europe. The main features of these two systems are given in Figure 19.7. As shown in the figure, the scanning parameters (scanlines/frame and refresh rate) are different for the two systems.
Standards Features Scanlines/frame Refresh (frames/second) Transmission Modulation Media
MUSE 1125 rate 60
HD-MAC 1250 50
Analog Analog FM FM Broadcasting Broadcasting satellite satellite Used in Japan Europe Figure 19.7. Features of the two HDTV standards. Apart from broadcasting purposes, HDTV is used in industrial applications also. Museums, film production, teaching, electronic billboards, and multimedia games are some such applications. Hardware Requirements for Video The following hardware devices are generally required in a computer system capable of handling video (see Chapter 9 for a description of some of these I/O devices): 1. A video camera is the most commonly used input device for capturing video data. 2. A video monitor (TV monitor or computer monitor) is the most commonly used output device for displaying video data. 3. A video board (or video card), which is equipped with A/D and D/A converters. In addition to performing the basic function of A/D and D/A conversions of video signals, a video board also has connectors for video camera and video monitor. Note that the video camera, the video monitor, and the video board must be compatible with each other. A video board also compresses the video image for more efficient storage. 4. Video editors are used to cut and paste video sequences, to add special effects, and to create new video sequences from existing video sequences. They also allow superimposing of text and audio on a video sequence. Software Requirements for Video In addition to the above listed devices, the following capabilities are highly desirable in a multimedia computer system with video facility: 1. Video clips. This is a library of video clips from which one can select and directly import a video clip and use it in a multimedia application. This saves enormous time and effort that might otherwise be required to create a similar video clip. A video clip library often provides the facility to add a new video clip or delete an existing video clip from the library.
2. Recording and playback capabilities. A multimedia system with video capability should allow the user to control the recording and display of a video sequence. For example, it should provide the user with options to 'pause' and 'replay' the animation sequence. MULTIMEDIA APPLICATIONS It was mentioned in the beginning of this chapter that the capability of computers to handle different types of media makes them suitable for a very wide range of applications and users. Below we discuss some of the most commonly known applications of multimedia. Please note that there are many more multimedia applications in addition to those discussed here and it is not possible to cover all of them. The basic idea here is to introduce the readers to some key multimedia applications so that they can gain sufficient background to appreciate the use of multimedia and to visualize its other applications. For ease of presentation, the applications discussed below have been grouped in areas. Education Education is an area that has benefited a lot with the advent of multimedia. There are many multimedia applications that directly or indirectly contribute to better education mechanisms. Some of the main applications are described below. Multimedia Books Multimedia books use audio, animation and video to enrich text and still images used in printed books to provide better understanding of the subject matter covered in a book. These books also provide a very effective way of searching for specific facts in a variety of ways such as searching mechanisms for words and phrases and hypertext and hypermedia techniques. Due to these reasons, several reference books such as encyclopedias, dictionaries, telephone directories, etc. are now being published and made available on CD-ROMs, which allow users to access the information they need quickly. Many journals, magazines and product manuals are also being published in electronic form so that users can access and print only those articles/portions in which they are interested. A further extension of this idea is digital library, which is a repository of large information in digital form allowing multiple users to search and access information of their interest from it. Such a library (being virtual nature because it stores the books only in soft copy form and not physically) can remain open round the clock allowing users to access information from it at their own convenient time. Network multimedia can be used to allow users to access the library from remote locations. With this extended facility, there is no need for a user to
physically come to the library for using it. This overcomes the distance and time limitations of current libraries. Better Presentation We saw that multimedia applications can present information in a variety of forms (text, graphics, animation, audio, video) involving the use of more of our senses. Hence multimedia presentation can be used to better explain a subject matter to the students because it enhances the comprehension capability of the students. It is extremely effective in getting across new ideas and concepts. Moreover, it helps in making a presentation more interesting and has the inevitable effect of grabbing and holding a person's attention. Foreign Language Learning Using a book for learning a foreign language is difficult because one is not sure how to pronounce the words written in text form. The book can have an accompanying audio tape to solve this problem, but it is inconvenient for the learner to keep rewinding the tape to find specific information. Moreover, the learner also does not have the flexibility to quickly hear the pronunciation for a specific word. Using a multimedia program incorporating both text and sound, the learner can see words and phrases on the screen as they are pronounced by the computer program. The learner also has the flexibility to request the computer to pronounce any word displayed on the screen by simply selecting that word (say by clicking over it). Additional features may be incorporated in the multimedia program to assist in faster learning process. For example, if the learner does not understand the meaning of a word or phrase, he/she may point to it to quickly display a translation for it in his/her native language. Natural Learning Environment Multimedia course materials provide a natural learning environment because they allow different students to learn and proceed at their own pace of learning. For example, beginners may like to spend more time on a subject matter to understand the basic concepts, while advanced learners may like to skip rudimentary information and directly go to more advanced concepts. Hypertext and hypermedia technologies may be used in multimedia course materials to provide this flexibility to the students. With this facility, since the students are not forced to read through the subject matter in a rigid format, they find the learning environment more interesting and motivating to learn faster. Networked multimedia can be used to simulate a virtual classroom in which students can come for learning at their own convenient time and can leave the
classroom whenever they want. Note that just like a digital library, a virtual classroom (which does not exist physically) overcomes the distance and time limitations of current classrooms because it remains open round the clock and allows students to use it from remote locations. Training Employee training is an important aspect of any organization. An organization often needs to impart training to its employees on the processes, products and services offered by the organization. Many organizations have developed and use interactive multimedia training programs for this purpose. A computer based training has many advantages. It allows the employees to take training at their own convenient time. For example, workers on a nigh shift can learn as conveniently as those working during the day, and an employee can stop training when other more important work must be done, and resume training when his/her schedule allows. Multimedia tools are often used in these training programs to enhance their effectiveness. For example, hypertext and hypermedia technologies may be used to allow different employees to get trained at their own pace of learning. Networked multimedia can be used to impart the same (high quality) training to employees working in remote offices and plants of the organization. Entertainment Computer systems were normally considered as a device that is useful in our education, research, and work. However, multimedia technology has now also made them an interesting entertainment tool. The applications described below are few examples of this. Video Games With the advent of multimedia, there has been a boom in the different types of video games available on computer systems. Sound and animation have been particularly important in these applications to make the games thrilling and exciting for the user. Because the sequence of events in such games depends on the actions of the player, these programs are very interactive in nature and offer an ever-changing contest to the user. The virtual reality technology has helped in creating new types of video games that were never possible before. For example, Virtual World Entertainment makes the Battletech Center, in which players sit in a tiny booth where they are fully immersed in a virtual battle. Players can fight with the computer, or they can be networked together to play against each other.
Special Effects in Films Multimedia technology is also being exploited by the film industry for providing special effects in films. Several movies now contain many visual tricks that could never be accomplished without the aid of computers. For example in the Hindi movie, Chachi 420, the hero is shown being transformed from man to woman by the use of a computer graphics technique called morphing. In the English movie, Jurassic Park, the dinosaurs were shown as living creatures (although they are extinct and no more exist) by the use of multimedia technology only. Similarly, a number of very successful films, including Titanic, Star Wars, Superman, have used multimedia technology for special effects. Animation Films Computerized animation and coloring have given a new life to the animation film industry. In animation films of the past, animations were created by using handdrawing technique, which was a very tedious and time consuming process. Because of hand-drawing, the quality of animations was also not very good (was far from realistic). Using computer programs based on the precise mathematical laws of physics, animators can now create animations that are far more realistic than ever before. Moreover, with computerized animation facility, an animator can now create in a few hours what used to take days with hand-drawing techniques. Elaborate scenes can now be animated in much less time and cost that might once have been too costly to bring to life, such as the flying-carpet ride in Disney's Alladin. Interactive TV Interactive TV allows users to select TV programs of their interest from remote massive video archives, view at the time they wish without leaving the comfort of their homes, and interact with the programs via VCR-like functions, such as fastforward and rewind. In essence, it implements a virtual video-rental shop from which customers can borrow and view the videos of their choice at their own convenient time. The rental shop being virtual provides 24 hours service. Interactive TV is implemented by using a Video-on-Demand (VoD) system and the facilities offered are often referred to as movie-on-demand, news-on-demand, sports-on-demand, education-on-demand, etc. depending on which ones of these are supported by the system. The main components of a VoD system are as follows (see Figure 19.8): 1. Video server. A video server consists of massive storage and media controllers. It stores a large number of digitized videos and serves simultaneous
video requests to the same or different videos on demand. The storage media usually consist of magnetic disks, optical disks, and magnetic tape and are usually organized hierarchically for cost effectiveness. Under this configuration, popular videos are stored in the disks. Less popular ones are stored in tape devices with terabyte capacity and retrieved as necessary to the disk drive for processing. The basic functions supported by video servers include request handling, random access, and user interactions. 2. Transport network. The transport network delivers video programs from the video server to the customers. The network must have a very high data rate and satisfy the real-time delivery constraints of video traffic. 3. Set-top box. A set-top box along with the television monitor and the infrared controller (i.e., remote control), serves as the bridge between the subscribers and the system. With on-screen images and cursor-like devices, viewers are connected to the video server and can browse the video selections and interact with the services via VCR-like functions. The major functions of the set-top box include receiving the incoming video streams; demodulating, demultiplexing, and decoding the signals; performing the necessary signal conversion, such as D/A transformation for playback on the TV monitor; and sending outgoing control messages. Marketing In the last few decades, computers have radically altered business practices. Because multimedia is so effective at communicating information, many companies have found it to be a powerful marketing tool. Few examples of use of multimedia in marketing are given below. Multimedia Presentation Business promotion often requires making presentations by a company to its prospective customers about the company and its products and services. Such presentations are normally carried out either at the company's own premises, or at the premises of a specific customer, or at an exhibition/conference site. Multimedia computer systems are nowadays preferred and being used extensively for this purpose instead of slides or overheads. The main reason is that the audiovisual and other features of multimedia make the presentations more lively and interesting to the audience. Moreover, changes can be easily incorporated in an already prepared presentation to customize it for a specific customer or a specific group of prospective customers. Laptop computers have further helped in this activity and are extensively being used nowadays to carry out multimedia presentations at locations other than the company's own premises because marketing agents need mobility.
Multimedia Catalogue Multimedia brochures and catalogues are extensively used nowadays for sales promotion and marketing. For example, many real estate firms use multimedia to show prospective customers information about available houses. Such application can quickly search a few houses from a database of many houses that closely match a customer's requirements. It can then display high-quality graphic images of each selected house, with room photographs, layouts, and maps to facilitate the customer in better and faster decision making. Although the same information can be provided on paper, the computerized application has the added capabilities of rapid searches to locate houses that meet the needs of the customer. Multimedia Kiosk Multimedia kiosks are often being used nowadays in public places as information providers in place of help desks with an attendant. For example, a multimedia kiosk placed at an airport might provide the travellers with information on tourist attractions, restaurants, and hotels, along with graphic images and maps showing how to gel there. A kiosk at a large museum may be used to provide the visitors with information on what types of displays are in which parts of the museum. A kiosk at a cosmetic counter in a supermarket may provide customers witr information on makeup for particular skin types and colors. A multimedia kiosk generally has a touch-screer monitor with a very user-friendly graphical user interface for ease of operation by general public. Animated Advertisements For sales promotion, companies often advertise their products and services on television. Most of the TV programmes are sponsored by one or more companies and in turn they get time slots for advertising their products during the programme. Several of these advertisements use computer animation to effectively communicate the usefulness of the products or services. For example, if you carefully watch the advertisements for detergent soaps/powders, toothpastes, insecticides for mosquitoes and cockroaches, etc., you will see few scenes in them, which cannot be videographed. These scenes are produced using computer animation. Virtual Shops and Shopping Malls The already described video-on-demand system is also used to create virtual shops, which remain open 24 hours and allow the customers to do purchases at the time they wish without leaving the comfort of their homes. In this case, the video server stores a large number of digitized videos containing multimedia description
of all the products on sale in the shop. A customer makes a request to the system for a particular type of product by selecting the product name from the displayed list of products and then providing its specifications as requested by the system by using the TV set at his/her home. The system then quickly searches for all the models of the named product in the shop that match the customer's requirements. It then displays the list of selected models and at allows the customer to interactively examine each model in the list by requesting more details of specific aspects of the model. For example, if the customer is interested in buying a fridge, the system displays a list of all models available, and then on customer's request displays a high quality graphic image of the selected model. The customer can then interactively rotate the image to examine the look of the model from all directions, can open the doors of different compartments one-by-one to examine the facilities inside, can zoom certain parts inside the fridge to have a closer look, can request for display of the image in all colors in which the model is available, etc. Once the customer gets satisfied and wants to purchase one, he/she places an order by filling up a form displayed by the system. The form requests for information such as product name with model number and color, name a address of the customer, credit card number, delivery address (the person ordering may like to gift it to someone), choice of delivery date (for example on birthday), etc. The system records the order and the purchased item is delivered manually to the customer by the shop owner. The virtual reality technology is used to extend the concept of virtual shops to virtual shopping malls. In this case, the customer is presented with a virtual shopping street on his/her TV screen with many different types of shops lined up on both sides of the street. Using interactive TV features, the customer can use the remote control to have a virtual walkthrough on the shopping street, enter any shop of his/her choice, have a closer look at the products displayed in it, make purchases in the shop, then get back on the street, enter another shop and make purchases there, and so on. Notice that with this facility, a customer sitting in his house in India can virtually roam around in a shopping street in Japan and can make purchases from the shops in that street. Groupware Groupware, also known a computer-supported cooperative work (CSCW), refers to a system that provides an electronic shared workspace to users located far from each other with communication, collaboration, and coordination supports. Group communication support allows the users to cooperate regardless of time and location constraints by exchanging messages either synchronously or asynchronously with each other. Group collaboration support allows users to share information among groups for effective collaboration. Group collaboration support effectively manages the participating users so as to enhance communication and
collaboration among them. Multimedia has been helpful in making groupware more effective by allowing the exchange and sharing of multimedia information among participating users. The following are the three most important multimedia groupware applications. Multimedia E-mail We saw the usefulness of e-mail application in the previous chapter. Multimedia e-mail is an extension of text-based e-mail facility. It allows users located far from each other to asynchronously exchange messages that contain multimedia information. In fact, e-mail is the most widely used service over the Internet. MIME (Multipurpose Internet Mail Extensions) is the current standard for Internet e-mail. Unlike the older SMPT (Simple Mail Transfer Protocol) standard, which understands only ASCII characters (text), MIME specifies a set of encoding rules and header extensions that describe new type of contents (such as image, video, or audio) and the structure of the multimedia message to embed multimedia information. Collaborative Authorship Collaborative authorship applications deal with collaboratively editing and composing a (multimedia) document by a group of people. The participants can access the document either synchronously or asynchronously. Each participant works on a part of the document and the final document is the combination of all individual parts. Multimedia Conferencing Multimedia conferencing, also known as video conferencing, refers to a system that simulates face-to-face interactions among participating users located far from each other as if they were sitting and discussing in a single room. In this system, each participating user has a PC (or a workstation) on his/her working desk, which has a video camera and mike attached to it. The PCs of all these users are interconnected together through a high-speed network. The network is used to transmit audio, video and other data from one participant's PC to the PCs of other participants. A user willing to communicate with other users, talks in front of his/her PC. The audio-visual equipment attached to his/her PC capture the information that is then transmitted to the PCs of other users. The coordination support software coordinates the proceeds of the conference and resolves any conflicts when multiple users want to communicate simultaneously.
Science and Engineering The area of science and engineering has been benefiting from the use of computers ever since the early days of computers. With the advent of multimedia, several new applications of computers emerged in this area. Today computers are playing an ever-increasing role in the area of science and engineering. Few examples of use of multimedia in science and engineering are given below. Computer Aided Design (CAD) CAD is an important application of generative graphics. Nowadays CAD packages are available for designing almost anything - cars, aircrafts, ships, buildings, printed circuit boards, etc. In fact, knowledge of CAD is now required for many engineering design jobs because an engineer or architect designing a product can be far more productive with a computer than with pencil and paper. Use of a CAD system in designing an object, allows the designer to create an electronic model by describing all three dimensions of the object. The designer can display any view of the object on the computer screen to see the object from a different perspective. On paper, the designer needs to draw a different drawing for each view, and for any change in the design, he/she will have to redraw all affected views. Visualization Visualization deals with exploring data and information graphically. Many computer applications deal with very large data sets. If the data is represented numerically, it becomes difficult to find relationships between different parts of the data. But if the same data is represented graphically, this relationship can be visualized much easily. For example, if we have a table containing data, which shows how pressure changes with the change of temperature in a particular experiment, it will be much easier to visualize this relationship if the same data is represented in the form of an (x, y) graph. Special visualization tools are available that allow graphical representation of very large and complex data sets assisting scientist and engineers in the analysis of their problems. Visualization is not only restricted to scientific and engineering applications. For example, a pie chart can be used to better visualize the percentage of earnings of a company from its different activities, or a bar chart can be used to better represent the year-wise change in profit of a company over a period of time. Data Compression We saw that multimedia data are typically large in volume and often require realtime and continuous delivery and presentation. Hence, multimedia applications
require massive storage space, high data-transfer rate, and huge transmission bandwidth. Existing computers, storage devices, and networks are unable to satisfy these requirements. Hence, data compression techniques are used to reduce the volume of multimedia data in such a way that the original data can be reproduced with the desired quality level wherever required. The process of reducing the volume of data by applying a compression technique is called compression and the resulting data is called compressed data. The reverse process of reproducing the original data from compressed data is called decompression and the resulting data is called decompressed data. We will see below that the decompressed data may or may not be the same as the original data depending on the type of compression technique used. Types of Compression Techniques Compression techniques can be categorized based on the following considerations: 1. Lossless or lossy. If the decompressed data is the same as the original data, it is referred to as lossless compression, otherwise the compression is lossy because there is loss of originality in the reproduced data. Lossy compression techniques usually have higher compression ratios (larger reduction in the volume of data), but may introduce some distortion in the original quality of the data when decompressed. In case of lossy compression, the desired quality level of the decompressed data is application dependent. 2. Symmetrical or asymmetrical. In a compression technique, if the time required to compress and to decompress are roughly the same, it is referred to as symmetrical compression. In asymmetrical compression, the time taken for compression is usually much longer than decompression. Symmetrical techniques are used in those applications that require data to be digitized and compressed in real time, such as video conferencing. On the other hand, asymmetrical techniques are used in those applications in which the data are compressed once and decompressed several time, such as in video-on-demand applications. 3. Software or hardware. A compression technique may be implemented either in hardware or software. As compared to software codecs (COder and DECoder), hardware codecs offer better quality and performance. However, they are costlier and are available only on platforms with costly extended hardware. Software codecs can be ported on any platform ensuring wider consumer accessibility. Due to performance reason, hardware codecs are suitable for CPU-intensive processing and software codecs are suitable for less CPU-intensive processes.
Basic Compression Techniques Some basic compression techniques are described below. An actual compression technique for multimedia object is often a combination of one or more of these basic compression techniques. Lossless Techniques The following two are the commonly used lossless compression techniques: 1. Run-length coding. In this method, repeated symbols in a string are replaced with the symbol and the number of instances it is repeated. For example, "aaaabbcccccaaaaaababbbb" is expressed as "a4b2c5a6blalb4". 2. Variable-length coding. In general, coding schemes for a given character set use a fixed number of bits per character. For example, BCD, EBCDIC and ASCII coding schemes use a fixed number of bits per character. If it is known (through statistical studies) that the frequency of occurrence of different characters in a character set is different (for example we all agree that the character Z is not so frequently used as the character A in the character set of English alphabets), then a coding scheme that employs a variable number of bits per character may be used to obtain tighter packaging of data. A variable-length coding scheme is based on this concept. In such a coding scheme, fewer bits are used to represent the more frequently occurring characters than the less frequently occurring ones. Obviously, the most frequently occurring character would be represented with only one bit. Huffman coding is an example of this method. It is briefly described below with an example. To make the example simple and easy to understand, let us assume that our character set consists of only six characters: A, B, C, D, E, and F. A conventional coding scheme that uses a fixed number of bits per character would require three bits per character (because 22 < 6 < 23). Now suppose the relative frequency of occurrence of the characters in this character set is as shown in Figure 19.9. The figure also displays a Huffman coding of these six characters, which minimizes the total number of bits for characters appearing with the frequency shown. Note that the Huffman encoding scheme pays off only with uneven character distribution. For example, if all characters in Figure 19.9 were used equally often, the mean number of bits per character would be 3.33. This is worse than a fixedlength coding scheme in which the mean number of bits per character would be 3. For probabilities given in Figure 19.9, the average number of bits per character is 2.05. Character
Probability of Code Occurrence
A 0.40 0 B 0.29 10 C 0.20 110 D 0.08 1110 E 0.02 11111 F 0.01 11110 Figure 19.9. An example showing a Huffman code for a given character set with the probability of occurrence of the characters. Lossy Techniques The following two are the commonly used lossy compression techniques: 1. Predictive encoding. In many types of multimedia objects (like digital audio or video) adjacent samples or frames are generally similar to each other. The predictive encoding method takes advantage of this fact and stores only the initial sample and the difference values between every two adjacent samples for all the samples in the compressed form. The compressed data is decompressed by reproducing a sample from its previous sample and the difference value between the two samples. Since the size of the difference value between samples is usually much smaller than the size of: sample itself, the file size of a compressed data is generally much smaller than the uncompressed data file. Note that the sample may be a pixel, line, audio sample, or video frame. 2. Transform encoding. In this method, data is converted from one domain to another that is more convenient for compression. For example, data may be converted from the time domain to the frequency domain. DCT (Discrete Cosine Transform) encoding is the best example of this method. It is described below. DCT encoding transforms samples from the time domain to the frequency domin. The input data is fed as a two-dimensional block, typically with 8x8 pixels. DCT transforms an 8x8 block of pixel color values into an 8x8 matrix of spatial frequencies. The low frequencies (i.e., general information) are positioned near the upper left corner, and the higher frequencies (i.e., sharp edges) are positioned near the lower right corner. The coefficients of the output DCT matrix are divided by the corresponding entries of an 8x8 quantization table (Q-table). Each entry of the Q-table describes the quality factor of the corresponding DCT output coefficient. The lower the quality factor, the higher the magnitude of value. For example, for DCT compression, the lower frequencies are more important than the higher ones. Therefore, a Q-table with smaller numbers near the upper-left corner and larger ones near the lower-right corner is used. The division of the DCT output matrix by the Q-table causes the lower-right corner portion of the resultant matrix to the resultant matrix to become mostly zeros. As a result, the greater portion of the high-frequency information is discarded. The modification of the Q-table is one way to change the degree of compression. The entries in the Q-table are based on
eye masking experiments and are essentially an approximation to the best estimate of levels that provide just-noticeable distortion in the image. The DCT decoder performs the reverse process and converts the 64 coefficients back to pixel values. Since quantization is a lossy process, it may not reproduce the originals exactly. Some Standard Multimedia Compression Techniques Over the years, several standards have evolved for compression and decompression of multimedia objects. Some of the most commonly used standards are described below. In selecting the standards to be described here, we have taken care to select one each for still image, video and audio media. JPEG (Joint Photographic Experts Group) JPEG is the standard compression technique for still images. It is of reasonably low computational complexity, is capable of producing compressed images of high quality, and can provide both lossless and lossy compression of arbitrarily sized grayscale and color images. As will be clear from the explanation below, it is a hybrid of several basic compression methods. It supports the following four modes of encoding: 1. Sequential. In this mode, the image is encoded in the order in which it is scanned. 2. Progressive. In this mode, the image is encoded in multiple passes, so that the entire content can be visualized in a rough-to-clear process (i.e., refined during succeeding passes). This capability is convenient for browsing applications where a low-quality (low-resolution) image is more than adequate for things like scanning through the pages of a catalog, and when the desired page is reached, only that page can be subjected to subsequent passes to display it more clearly (in high-resolution). 3. Hierarchical. In this mode, the image is encoded at multiple resolutions to accommodate different types of displays. 4. Lossless. In this mode, the image is encoded in such a way that the original quality of the image can be fully restored. JPEG uses DCT encoding and quantization for the first three modes, and uses predictive encoding for the lossless mode. Since quantization is a lossy algorithm, the first three modes of encoding are lossy. The basic steps of lossy JPEG compression (shown in Figure 19.10) are as follows (the decoder performs the inverse operations):
1. If the image to be compressed is color, RGB color space is transformed to YUV color space and chrominance component (the color) is downsampled to exploit color redundancy. If the image is grayscale, this step is not needed. 2. The image is then divided into a series of blocks of 8 x 8 pixels. 3. DCT transformation is then applied to each such 8x8 block of pixels to exploit spatial redundancy. 4. The resulting DCT coefficient (the 64 frequency component in each block) are then scalar quantized using a quantization-table (O-table). 5. The resulting 2-D output of quantization is then ordered in a zigzag sequence, creating a bit stream with zero value coefficients close to the end, since the lowfrequency component (most likely nonzero) are placed in front of the highfrequency components (most likely zero). 6. Run-length encoding is then applied to the zeros of the zigzag sequence. 7. Finally, variable-length encoding is applied on the run-length coded stream to obtain the JPEG compressed stream. The lossless mode of JPEG compression uses a different method than the lossy method shown in Figure 19.10. Fundamentally, the image pixels are handled separately (i.e., the 8 x 8 block structure is not used), and each pixel is predicted based on three adjacent pixels using one of eight possible predictor modes. The basic steps of lossless JPEG compression (shown in Figure 19.11) are as follows (the decoder performs the inverse operations): 1. A simple predictive coding scheme is first applied to the image to be compressed. For this, each pixel of the image is handled separately. 2. A variable-length encoder is then used to encode the predicted pixels losslessly to obtain the JPEG compressed bit stream. MPEG Video MPEG (Motion Pictures Experts Group) is a multimedia standard with specifications for coding, compression, and transmission of audio, video, and data streams in a series of synchronized, mixed packets. MPEG video standard is a subset of the MPEG standard. It is divised for compression of a wide range of video and motion pictures. It defines the structure of coded video bit streams transmitted to the decoder and the decoder architecture, but leaves the encoder architecture undefined, so that some encoders may produce higher quality compression, some may be real-time, and some may require manual intervention. MPEG compression is lossy and asymmetric, with the encoding process requiring more time than the decoding process. Digital video compression may either apply intraframe compression to each individual frame of the video or combine both intraframe and interframe
compression. The former scheme yields a lower compression ratio than the latter because it does not exploit temporal redundancy. Differential encoding is one approach to remove temporal redundancy. It encodes and sends the differences in the adjacent frames. Motion compensation is another technique for interframe compression. It compares the current frame with the previous one and encodes the motion vectors (i.e., the change in coordinate values due to the motion) and pixel differences after motion. MPEG uses both intraframe and interframe techniques for data compression. In MPEG, encoded frames are of the following three type: 1. I (Intraframe). An I-frame is coded independently of any other frames in the video sequence. Hence it contains all of the necessary information required to reproduce a complete frame. It basically applies the intraframe compression of the basic JPEG DCT algorithm. 2. P (Predicted). A P-frame is predicted from the differences from the previous Ior P-frame in the sequence. 3. B (Bidirectional). A B-frame is bidirectional interpolated by the preceding and the subsequent I- 01 P-frames. Note that due to the interframe dependency in case of P-frames and B-frames, a current P-frame can be decoded only if the preceding I- or P-frame has been presented to be decoded. Similarly, without both the preceding am subsequent Ior P-frame in the decoder, it is impossible to generate a B-frame. For example, if the presentation order of a video sequence is IBBPBBPBBI.....then the actual input sequence of MPEG frames to a decoder should be IPBBPBBIBB.....(see Figure 19.12). Therefore, the patterns of video frame sequence in storage and during transmission are different from that for presentation. The basic steps of MPEG compression (shown in Figure 19.13) are as follows (the decoder performs the inverse operations): 1. The individual frames of the video data to be compressed are preprocessed and color subsampling is performed on them to carry out the first level of compression. Recall that color subsampling is approach shrinks the data size by downsampling the chrominance component (i.e., using less bits to represent the chrominance component) while leaving the luminance component unchanged. 2. Interframe motion compensation is then performed for the frames of types P and B to carry out the second level of compression. This step is not needed for the frames of type I because only intraframe compression is applied to these frames. 3. Each frame is then divided into a series of blocks of 8 x 8 pixels. 4. DCT transformation is then applied to each such 8x8 block of pixels to exploit spatial redundancy. 5. The resulting DCT coefficients (the 64 frequency components in each block) are then scalar quantized using a quantization table (Q-table).
6. The resulting 2-D output of quantization is then ordered in a zigzag sequence, creating a bit stream with zero value coefficients close to the end. 7. Run-length encoding is then applied to the zeros of the zigzag sequence. 8. Finally, variable-length encoding is applied on the run-length coded stream to obtain the MPEG compressed video stream. MPEG standard has the following versions: 1. MPEG-1. The MPEG-1 standard was developed in response to industry needs for an efficient way of storing and retrieving video information on digital storage media. One such inexpensive medium being the CD-ROM, the first version of MPEG, MPEG-1, was targeted at CD-ROM and applicatioins at a bit rate of about 1.5 Mb/s. Although the MPEG-1 standard was developed with CD-ROM in mind, other storage and transmission media can include DAT, Winchester disk, optical disk, ISDN and LAN. In fact, the data rate is variable and all decoders must be able to decode at rates up to 1.856 Mb/s. The video is strictly progressive (i.e., noninterlaced) and picture quality is equivalent to VHS. 2. MPEG-2. The MPEG-2 standard was developed to provide the capability of compressing, coding, and transmitting high-quality, multichannel, multimedia signals over broad-band networks. It addresses high-quality coding for all digital transmissions of broadcast-TV-quality video at data rates of 2-15 Mb/s. Its major applications include digital storage media, digital television (including HDTV), broadcasting over cable, satellite, and other broadcast channels, and other communications applications. Since MPEG-2 was designed mainly as a transmission standard, it supports a variety of packet formats (including long and variable-length packets of from 1 KB up to 64 KB) and provides error-correction capability that is suitable for transmission over cable and satellite links. MPEG-2 is backward compatible with MPEG-1. The primary enhancement is centered around the addition of interlaced video. Thus, there are two coding structures (field and frame) in MPEG-2. Other improvements include enhancements on coding efficiency. 3. MPEG-4. Recently, the focus of video coding has shifted to object-based coding at bit rates of 8 Kb/s or lower and 1 Mb/s or higher. Key aspects of this newly proposed MPEG-4 standard include independent coding of objects in a picture frame; the ability to interactively composite these objects into a scene at the display; the ability to combine graphics, animated objects, and natural objects in the scene; and the ability to transmit scenes in higher dimensionality formats (e.g., 3-D). Also inherent in the MPEG-4 standard is the concept of video scalability, both in the temporal and spatial domains, in order to effectively control the video bit rate at the transmitter, in the network, and at the receiver so as to match the available transmission and processing resources. MPEG-4 builds on and combines elements from three fields - digital television, interactive graphics, and the World Wide Web. It aims to provide a merging of the production, distribution, and display elements of these three fields.
MPEG-4 is now in the application-identification stage. It currently targets applications with very low bit rates of 48 - 64 Kb/s, such as videophone over the PSTN and visual transmission over mobile facilities. MPEG Audio MPEG audio is a standard for compression and decompression of digital audio. It has been found that whenever a strong signal is present, the spectral neighborhood of weaker audio signals becomes imperceptible to human ears. Measurements of human hearing sensitivity have shown that the ear integrates audio signals that are very close in frequency and treats them as a group. Only when the signals are sufficiently different in frequency will the era perceive the loudness of each signal individually. The coding technique used in MPEG audio standard (known as perceptual coding) takes advantage of this perceptual weakness of human ears (pshychoacoustic phenomena). In perceptual coding, the audio spectrum is divided into a set of a narrow frequency bands, called critical bands, to reflect the frequency selectivity of human hearing. Then a filtering mechanism is used to filter coding noise sharply to force it to stay close to the frequency of the frequency components of the audio signal b6ing coded thereby masking out the noise. By reducing the coding noise when no audio is present and allowing strong audio to mask the noise at other times, the sound fidelity of the original signals can be perceptually perceived. The basic steps of MPEG audio (MPEG-1 audio, to be more precise) encoding, shown in Figure 19.14(a), are as follows: 1. The audio signal to be compressed is first transformed from the time domain to the frequency domain. 2. The resultant audio spectrum is then passed through a filter bank, which divides it into multiple sub-bands. 3. The input audio signal is simultaneously subjected to a psychoacoustic model, which exploits auditory masking and determines the signal-to-noise ratio of each sub-band. 4. Each sub-band is then quantized using a quantizer. A bit/noise allocation algorithm is used to determine the number of code bits allocated to quantize each sub-band so that the audibility of quantization noise is minimized. 5. The quantized samples are then formatted into a series of bit streams, which is the encoded bit stream for the input audio signal. As shown in Figure 19.14(b), the decoder performs the inverse operation of the encoder for reconstructing the time-domain audio signal from the encoded bit stream. To provide different quality and different bit rates, the standard supports three distinct layers for compression layer I is the most basic and simplest coding algorithm, and layers II and III are enhancements of layer I.
Recently, a new version of MPEG audio (MPEG-2 audio) has been developed for low-bit-rate coding of multichannel audio. It has 5.1 channels of audio, including three front channels (i.e., left, center, right) plus two surrounds, and a lowfrequency bass-effect channel called the subwoofer. Since the subwoofer has only limited frequency response (about 100 Hz), it is sometimes referred to as the ".1" channel. MPEG-2 audio is backward compatible to MPEG-1 audio because it uses the same coding methods and syntax as that of MPEG-1 audio for the two main channels (left and right). However, MPEG-2 audio uses new coding methods and syntax for the surround channels. MULTIMEDIA SYNCHRONIZATION Often multimedia presentation involves simultaneous presentation of two or more types of media. For example, a movie involves simultaneous presentation of video and audio. Similarly, a French movie with English subtitles involves simultaneous presentation of video, audio and text. In such cases, multimedia synchronization is required to ensure the desired temporal ordering among a set of multimedia objects in a multimedia presentation. One such example is lip synchronization, which refers to the requirement that the voice should match the movement of the lips of the speaker. Another example is subtitle synchronization, which refers to the requirement, that the timing of the displayed text (subtitle) should match with the timing of the video (scene) and the voice (spoken words) of the speaker that correspond to the subtitles. Multimedia synchronization can be divided into two levels - temporal specification and temporal synchronization. They are described below. Temporal Specification Temporal specification is used to model the temporal relationships among the set of related abstract multimedia objects of a presentation scenario, where an abstract object denotes a logical unit of media data such as a video object or a text object. The two commonly used approaches for temporal specification are graphicalbased and language-based. These are described below. Graphical-based Approach In this approach, the temporal relationships among the multimedia objects of a presentation scenario is represented in the form of a graph, such as an X-Y graph with X-coordinate representing time and Y-coordinate representing objects, or a Petri net. For example, consider the multimedia scenario represented as an X-Y graph in Figure 19.15.
It has four multimedia objects 0h 02, 03 and 04. St denotes the start point of object 0i. The same multimedia scenario is represented in Figure 19.16 as a petri net. Language-based Approach In this approach, the temporal relationships among the multimedia objects of a presentation scenario is represented in the form of a special programming language that has temporal synchronization functions. A few examples of synchronization functions that such a language may have are: 1. SyncAtomicSerial (A, B): Object B starts after object A stops. 2. SyncAtomicParallel (A, B): Objects A and B start simultaneously. 3. SynElemSerial (tl, A, t2, B): Object A starts after time tl, and object B starts after time t2 after A stops. 4. SyncElemParallel (tl, A, t2, B): Object A starts after time tl, and object B starts after time t2. Based on the values of tl and t2, objects A and B may overlap. Temporal Synchronization Note that temporal specification focuses only on the temporal relationships among the objects without explicit reference to time. Temporal synchronization mechanisms use the temporal relationships information of temporal specification and convert it into a set of playback schedules and retrieval schedules to enforce the playback deadlines for various multimedia objects of a presentation scenario. A playback schedule is the set of playback deadlines of multimedia objects in a multimedia scenario, where playback deadline of an object is the latest starting time of the object, for a meaningful presentation. Retrieval schedule is the set of request deadlines to the server. It accounts for end-to-end delays and ensures that the playback deadline is not missed. To display an object O at time Ts (the playback deadline), the presentation site sends a request to the corresponding media server at request time Tr, which is R time units prior to Ts. R is the response time to account for possible experienced end-to-end delays and packet losses. The timing relationships among the playback deadline Ts, the request time Tr, and the response time R are given by Tr = Ts - R and are depicted in Figure 19.17. The possible experienced end-to-end delays are contributed by (assuming that the presentation and server sites are different): 1. Request queuing delay and retrieval delay (i.e., seek, latency, and data transfer times) at server site. 2. Round-trip queuing delays, propagation delays, and transmission delays in network communication. 3. Queuing delay and processing delay at the presentation site.
The time duration from when a user initiates a presentation request to when the presentation begins is called the set-up time. Figure 19.18 shows the set-up time in a presentation. The set-up time T must be long enough to compensate for the system delays and to achieve inter-media synchronization. The set-up time therefore is not necessarily equal to max {R!t R2, —, Rn} where Oh O2, —, On are the start objects and Ru R2, — Rn are their respective response times. T is largely dependent on the request time of the first request objects in the request ordering. For example, in Figure 19.18, T' is largely dependent on the request times of objects O1 and O2. The set-up time determines the earliest playback time of the corresponding scenario, which in turn determines the absolute playback deadlines of the subsequent objects. The earliest playback time is the earliest time a presentation can start after it is initiated, and it is also the actual playback deadline of the start objects. Combining the earliest playback time with the relative playback schedule gives the actual playback schedule. Points to Remember 1. Multimedia refers to the use of more than one of the following media for information presentation: • Text (alphanumeric characters) • Graphics (line drawings and images) • Animation (moving images) • Audio (sound) • Video (videographed real-life events) 2. A multimedia computer system is a computer system that has the capability to integrate two or more types of media (text, graphics, animation, audio, and video) for the purpose of generation, storage, representation, manipulation and access of multimedia information. 3. In general, the data size for multimedia information is much larger than plain text information. Because of this, multimedia computer systems require: • Faster CPU (for more quickly processing large amount of data), • Larger storage devices (for storing large data files), • Larger main memory (for running programs with large data size), • Good graphics terminals (for displaying graphics, animation and video), and • I/O devices required to play any audio associated with a multimedia application program. 4. Keyboards, OCRs, computer screens, and. printers are some of the commonly used hardware devices for processing text media. 5. Text editing, text searching, hypertext, and text importing/exporting are some of the highly desirable features of a multimedia computer system for better presentation and use of text information.
6. Computer graphics deals with the generation, representation, manipulation and display of pictures (line drawings and images) with the aid of a computer. 7. Locating devices (such as a mouse, a joystick, or a stylus), digitizers, scanners, digital cameras, computer screens with graphics display capability, laser printers, and plotters are some of the commonly used hardware devices for processing graphics media. 8. Some of the highly desirable features of a multimedia computer system for better presentation and use of graphics information are painting or drawing software, screen capture software, clip art, graphics, importing, and software support for high resolution. 9. Computer animation deals with the generation, sequencing, and display (at a specified rate) of a set of images (called frames) to create an effect of visual change or motion, similar to a movie film (video). Animation is commonly used in those instances where videography is not possible or animation can better illustrate the concept than video. 10. Virtual reality is a relatively new computer technology using which the user can put a pair of goggles and glove and tour a three-dimensional world that exists only in the computer, but appears realistic to the user. 11. Scanners, digital cameras, video capture board interfaced to a video camera or VCR, computer monitors with image display capability, and graphics accelerator board are some of the commonly used hardware devices for processing animation media. 12. Some of the desirable features of a multimedia computer system with animation facility are animation creation software, screen capture software, animation clips, animation file importing, software support for high resolution, recording and playback capabilities, and transition effects. ......_ 13. Computer audio deals with synthesizing, recording, and playback of audio or sound with the aid of a computer. 14. Transducer is a device capable of changing signals from analog form to digital form or vice-versa. 15. MIDI (Musical Instrument Digital Interface) is an international standard developed for digital music that determines the cabling, hardware, and communications protocol needed to connect a computer with electronic musical instruments and recording equipment. 16. Sound board, microphone, speaker, MIDI devices, sound synthesizer, sound editor and audio mixer are some of the commonly used hardware devices for processing audio media. 17. Some of the desirable features of a multimedia computer system with audio facility are audio clips, audio file importing, software support for high quality sound, recording and playback capabilities, text-to-speech conversion software, speech-to-text conversion software, and voice recognition software.
18. Computer video deals with the recording and display of a sequence of images at a reasonable speed to create an impression of movement. Each individual image of such a sequence is called a. frame. 19. The three main TV broadcasting standard are NTSC, PAL and SECAM. 20. Video camera, video monitor, video board, and video editor are some of the commonly used hardware devices for processing video mediae 21. Some of the desirable features of a multimedia computer system with video facility are video clips and recording and playback capabilities. 22. Some of the key areas that have benefited a lot with the advent of multimedia are education, entertainment, marketing, groupware, and science and engineering. 23. Data compression techniques are used to reduce the volume of multimedia data in such a way that the original data can be reproduced with the desired quality level whenever required. 24. The process of reducing the volume of data by applying a compression technique is called compression and the resulting data is called compressed data. The reverse process of reproducing the original data from compressed data is called decompression and the resulting data is called decompressed data. 25. Data compression techniques may be lossless or lossy, symmetrical or asymmetrical, and software 01 hardware implemented. 26. The commonly used lossless compression techniques are run-length coding and variable-length coding. 27. The commonly used lossy compression techniques are predictive encoding and transform encoding. 28. Some standard multimedia compression techniques are • JPEG (Joint Photographic Experts Group) for still images • MPEG (Motion Pictures Experts Group) video for digital video • MPEG audio for digital audio 29. Multimedia synchronization deals with mechanisms to ensure the desired temporal ordering among a set c multimedia objects in a multimedia presentation. Questions 1. What is multimedia? Explain with the help of some examples. 2. Tell which ones of the following are unimedia and which ones are multimedia devices (give justification f< your answers): (a) TV system (d) A book having only text (no diagrams) (b) Stereo system (e) A book having both text and diagrams (c) Telephone system (f) A modern personal computer 3. List out the commonly used media types generally used in modern computer systems and explain how each one helps in information presentation. 4. What is a multimedia computer system? What are it typical characteristics? Do all multimedia systems need to have all these characteristics?
5. You have a printed document of 100 pages. Give two different ways in which this text data can be input to a computer system. 6. Name two output devices that can be used with a computer system to output text information. Give their relative advantages and disadvantages. 7. List some of the desirable features of a word processing software. Give a practical use of each feature. 8. What is meant by "text style"? List some of the text styles that you will like to have in your computer system. Give a practical use of each of them. 9. Explain with three examples how text searching feature can be useful in a computer system. 10. What is a "hypertext?" Explain how it helps in presentation and use of information. 11. What is meant by "text importing" and "text exporting"? Explain how these features help in the creation of a document. 12. Write short notes on: (a) Text editor (b) Hypertext (c) Text style (d) Text searching (e) Multimedia computer systems (f) I/O devices for text processing (g) Text importing and exporting 13. What is computer graphics? Give a few examples to explain its uses in multimedia applications. 14. Explain the difference between "generative graphics" and "cognitive graphics". Give two uses of each. 15. What is a "pixel"? Explain how an image is composed and displayed on a computer screen. 16. An image has a resolution of 1024 x 768 and color depth of 24-b (i.e., 8 bit each for the R, G, and B components). Calculate the storage requirement of this image. 17. Name two input devices for generative graphics applications. Describe a practical use of each. 18. Name two input devices for cognitive graphics applications. Describe a practical use of each. 19. Name three output devices that may be used with a computer system to output graphics data. Describe a practical use of each of these devices. 20. Explain with suitable examples the difference between "Standard VGA" and "Extended Graphics" displays. 21. What is a painting/drawing software? Describe two uses of this software in creating multimedia applications. 22. What is a screen capture software? Describe a typical use of this software.
23. What is a clip ait? How does it help in the creation of multimedia applications? 24. What is meant by "graphics importing"? Explain how this feature helps in the creation of multimedia applications. 25. Name some of the common graphics file formats. 26. Write short note on: (a) Locating devices (b) Scanner (c) Digitizer (d) Plotter (e) Clip art (f) Graphics importing (g) Standard VGA display (h) Extended graphics (i) Painting/drawing software (j) Screen capture software (k) Software support for high-resolution graphics 27. What is computer animation? Give a few examples to explain its uses in multimedia applications. 28. Differentiate between "animation" and "video". Explain the difference with the help of few examples. 29. What is "virtual reality"? Describe a multimedia application that makes use of this technology. 30. A jerk-free full motion animation sequence is created of a set of images having a resolution of 640 x 480 and color depth of 24-b (i.e., 8 bits each for the R, G, and B components). Calculate the storage requirement for storing one second of this animation. 31. What is an animation creation software? Describe with the help of a typical example how this software can be used to create an animation sequence. 32. What is an animation clips library? How does it help in the creation of multimedia applications? 33. Name some of the common animation file formats. 34. Name some of the transition effects features that may be used to make animation more interesting. 35. Write short notes on: (a) Refresh rate (b) Virtual reality (c) Graphics accelerator board (d) Animation creation software 36. What is computer audio? Give a few examples to explain its use in multimedia applications. 37. What is a transducer? Name three devices that can be categorized as a transducer.
38. Explain the role of analog-to-digital and digital-to-analog converters in the acquisition and playback of analog signals by a computer system. 39. What is a sound board card in a multimedia computer system? In addition to it basic function, what additional facilitiesdoes it generally provide? 40. What is MIDI? How does it help in the creation of multimedia applications? 41. Explain how an analog signal is transformed into a digital signal. Explain the role of sampling rate and quantization on the quality of the digital signal produced by this process. 42. An audio clip is of 30 seconds duration and uses 8 KHz sampling rate, 8 bits quantization, and mono recording. Calculate its storage requirement. 43. An audio clip is of 15 seconds duration and uses 44.1 KHz sampling rate, 16 bits quantization, and stereo recording. Calculate its storage requirement. 44. What will roughly be the storage requirement for (a) a 15 seconds telephone-quality audio clip (b) a 15 seconds CD-quality audio clip .. 45. What is an audio clips library? How does it help in the creation of multimedia applications? 46. Name some of the common audio file formats. 47. Write short notes on: (a) Transducer (b) Sound card (c) MIDI (d) Sampling rate (e) Quantization (f) Quantization error (g) Software support for high quality sound 48. What is computer video? Give a few examples to explain its uses in multimedia applications. 49. Give examples of two transducers that deal with video data. 50. Give examples of two transducers that deal with audio data. 51. What is the role of a video board card in a multimedia computer system? 52. Describe the process of displaying a video on a monitor screen. 53. Differentiate between "progressive scan" and "interlaced scan". 54. What are the main differences between a TV monitor and a computer monitor? 55. Define the following terms in context of video display: (a) Raster scanning (f) Progressive scan (b) Refresh rate (g) Interlaced scan (c) Scanline (h) Vertical blanking interval (d) Field (i) Horizontal blanking interval (e) Aspect ratio
56. Explain the difference between composite video signals and component video signals. Which of the two has better quality? Give one example for each of the following: (a) Analog video using composite video signals (b) Digital video using composite video signals (c) Analog video using component video signals (d) Digital video using component video signals 57. Name the three popular TV broadcasting standards. Describe their characteristic features. 58. What is HDTV? Name two standards that are based on this technology. List out the main characteristic feature of these two standards. 59. What is a video clips library? How does it help in the creation of multimedia applications? 60. Write short notes on: (a) Video display technology (b) Types of video signals (c) Broadcast TV standards (d) HDTV 61. Give some examples of multimedia applications in education. 62. What is a multimedia book? What are its main features? How do these features make it better than a conventional book? 63. What is a digital library? What are it advantages and disadvantages as compared to a conventional library? 64. Explain how is multimedia useful in (a) Better presentation (c) Learning a subject (b) Foreign language learning (d) Employee training 65. What is a virtual classroom? What are it advantages and disadvantage as compared to a conventional classroom? 66. Give some examples of multimedia applications in entertainment. 67. Explain how is multimedia useful in making (a) Video games (c) Animation film (b) Movies (d) Advertisements 68. What is an interactive TV? How is it different from a conventional TV system? 69. What is a video-on-demand system? What are its main components? What is the role of each component in the system? Name some applications of this system. 70. Give some examples of multimedia applications in marketing. 71. What is multimedia catalogue? What are its advantages and disadvantage as compared to a conventional catalogue? 72. What is a virtual shop? How can a customer make purchases from it? What are its advantages and disadvantage as compared to a conventional shop?
73. What is a virtual shopping mall? What are the key technologies involved in implementing it? 74. What is groupware? Give some examples of multimedia applications in groupware. 75. What is video conferencing? Explain how it works. 76. List out some of the uses of multimedia in scientific and engineering applications. 77. What is computer aided design? How does it help a designer? 78. What is visualization? Give two examples to illustrate its usefulness. 79. Write short notes on (a) Multimedia books (b) Digital library (c) Virtual classroom (d) Interactive TV (e) Video-on-demand system (f) Multimedia catalogue (g) Multimedia Kiosk (h) Virtual shop (i) Groupware (j) Multimedia e-mail (k) Video conferencing (1) Computer aided design (m) Visualization 80. What is data compression? Why is it needed in a multimedia computer system? 81. When an encoded data is decoded, is the resulting data same as the original data? Give reasons for your answer. 82. Explain the difference between lossless and lossy data compression techniques. Give one example each for justifying the use of one technique over the other technique. 83. Explain the difference between symmetrical and asymmetrical data compression techniques. Give one example each for justifying the use of one technique over the other technique. 84. Give the relative advantages and disadvantages of hardware and software codecs.. 85. Name two lossless and two lossy data compression techniques. Give a practical use of each of these techniques. 86. Write short notes on: (a) Run-length coding (b) Variable-length coding (c) Predictive coding (d) Transform coding (e) DCT coding
87. What is JPEG? Briefly describe the various modes supported by it. 88. Describe the steps involved in lossy JPEG compression. 89. Describe the steps involved in lossless JPEG compression. 90. What is MPEG video? Briefly describe its main characteristic features. 91. Differentiate among the different types of encoded frames used in MPEG video. 92. Describe the steps involved in MPEG video compression. 93. Write short notes on: (a) MPEG-1 video (b) MPEG-2 video (c) MPEG-4 video (d) MPEG-2 audio 94. What is MPEG audio? Briefly describe the basic principle on which it works. 95. Describe the steps involved in MPEG audio encoding. 96. Describe the steps involved in MPEG audio decoding. 97. What is meant by multimedia synchronization? Explain the need for it in multimedia applications by giving suitable examples. 98. Describe the roles of temporal specification and temporal synchronization in multimedia synchronization. 99. Describe with suitable examples two approaches that may be used for temporal specification in multimedia synchronization. 100. Define the following terms in context of temporal synchronization of multimedia objects: (a) Playback schedule (e) Response time (b) Playback deadline (f) Set-up time (c) Retrieval schedule (g) Earliest playback time (d) Request time 101. What is the relationship among playback deadline, request time, and response time in context of multimedia synchronization? 102. Explain how the actual playback schedule of a set of multimedia objects can be obtained from their temporal specification. 103. Write the full form of the following: (a) MPC (i) HDTV (b) CAD (j) MPEG (c) NTSC (k) MIDI (d) MUSE (1) VGA (e) JPEG (m) SECAM (f) OCR (n) HD-MAC (g) CAM (o) VOD (h) PAL
Chapter 20 Classification of Computers General purpose computers come in many sizes and capabilities. Traditionally, computers were classified by their size, processing speed and cost. Based on these factors, computers were classified as microcomputers, minicomputers, mainframes and supercomputers. However, with the rapidly changing technology, this classification is no more relevant. The problem is that computer technology is changing so fast that just after few months of introduction of a new computer in the market, new models of computers are introduced which have much higher performance and cost less. Thus, a recently introduced small system can outperform the large models of a few years ago, and a new minicomputer can do the work of an earlier mainframe at a much lower cost. Hence today computers are classified based on their mode of use. According to this classification scheme computers are classified as notebook computers, personal computers, workstations, mainframe systems, supercomputers, and clients and servers. In this chapter you will learn about these types of computers. NOTEBOOK COMPUTERS Notebook computers are portable computers which are mainly meant for use by people who need computing power wherever they go. As their name implies, notebook computers are approximately of the size of an 8½ x 11 inch notebook and can easily fit inside a briefcase. Since they have to be carried along, they are also light in weight, weighing around 2 Kg. They are also known as laptop PCs (laptop personal computers) because they are as powerful as a PC and their size and weight allows them to be used by comfortably placing them on ones lap. As shown in Figure 20.1, a notebook computer uses an almost full-size keyboard, a small flat screen liquid crystal color display, and a trackball (instead of a mouse because notebook computers are often used without a desk). They also have a hard disk (of few hundred mega bytes), a floppy disk drive, and sometimes a CD-ROM drive also. The display screen is foldable in such a way that when riot in use, it can be folded to flush with the keyboard to convert the system into notebook form. When in use, the display screen is folded open as shown in the figure. Many models of notebook computers can be plugged into a "dock" on a docking station (a personal computer or a workstation) to take advantage of the larger machine's big monitor, storage space, and other peripherals such as a printer. Many models of notebook computers can also be connected to a network so that they can download data (read files) from other computers on the network as and when such a need arises, or to access the Internet. Since notebook computers are meant to be
mobile and used from anywhere, efforts are being made to provide wireless connectivity to these systems with other stationary computers. Notebook computers are designed to be used even at places where there is no power point available to connect them with a power source (for example while traveling in a train or aeroplane). Hence they are designed to operate with chargeable batteries. With a fully charged battery, a notebook computer can be used for a few hours. Notebook computers normally run MS-DOS or WINDOWS operating system. They are mostly used for word processing, spreadsheet computing, data entry, and preparing presentation materials while a person is traveling. They are also used for making presentations by plugging them into an LCD (liquid crystal display) projection system when presentations are to be made at a location away from ones office. The processing capability of a notebook computer is normally as good as an ordinary PC (personal computer) because both use the same type of processor such as an Intel Pentium processor. However, a notebook computer generally has lesser hard disk storage than a PC to keep its total weight to around 2 Kg. Notebook computers are typically more expensive (2 to 3 times) than a normal PC. PERSONAL COMPUTERS (PCs) A PC is a non-portable general-purpose computer that can easily fit on a normal size office table (leaving some writing space to keep writing pads and other office stationary) and is generally designed to be used by one person at a time (singleuser-oriented). As the name implies, PCs were mainly designed to meet the personal computing needs of individuals either in their working places or at their homes. In fact, PCs have changed the work culture and work habits of numerous organizations and individuals. An increasing proportion of office work now involves the use of computers. Hence, today PCs are found on the working desks of several employees in any organization. PCs are also providing employees with flexible working environments. Those employees who could not work during traditional office hours due to personal reasons can now work part of the time in the office and the remainder of the time at home by having a PC in their homes. Several individuals also keep a PC in their homes to run a business in their homes. PCs are also used both by children and adults for education and entertainment. Hence, PCs are now very common everywhere and can be found in offices, classrooms, homes, hospitals, shops, clinics, etc. The configuration of PCs varies from one PC to another depending on their usage. However, the most commonly used configuration consists of a system unit, a
monitor (display screen), a keyboard, and a mouse. The system unit, which is in the form of a box, consists of the main circuit board (consisting of CPU, memory, etc.), the hard disk storage, the floppy disk drive, the CD-ROM drive, any special add-on cards (such as network interface card), and ports for connecting peripheral devices (such as printer). The two most commonly used models of PCs are the desktop model and the tower model. As shown in Figure 20.2, the only difference between the two models is that the monitor is positioned on top of the system unit in the desktop model, whereas the system unit of the tower model is designed to stand by the side of the monitor, allowing it to be positioned even on the floor beside or under the working desk to preserve desk space. Although the desktop model was more popular few years ago, the tower model is gaining popularity now. A PC generally employs several chips (CPU chip, RAM chips, ROM chips, I/O handling chips, etc.) on a main circuit board called a system board, or motherboard. The motherboard is what distinguishes one PC from another. Often PCs are distinguished by the main component of the motherboard, that is the microprocessor chip, which is used as their CPU. Some of the commonly known PCs and the microprocessor chips used in them are described below. PCs were first introduced in mid-1970s, and all PCs built during the 1970s were 8bit systems. That is, they were built using microprocessor chips that could operate on 8 bits of data at a time. Some popular microprocessor chips of this type are Zilog's Z80, MOS Technology's 6502, Intel's 8080, and Motorola's 6809. All these microprocessor chips have built-in eight-line data buses. This means that these chips can only retrieve from storage, manipulate, and process a single 8-bit byte of data at a time. They are thus 8/8 bit chips. A 16-line address bus is also built into these chips to determine the primary storage locations of the instructions and data that are needed. With 16 address lines, a maximum of 64 Kilobytes (2 16) or 65,536 separate storage locations can be identified by these chips. Some of the PCs that were built by using these types of microprocessor chips are Apple II, Commodore 64, and Radio Shack TRS-80 systems. In early 1980s, microprocessor suppliers introduced 16-bit microprocessor chips that could operate on 16 bits of data at a time. Intel 8088 is the most popular microprocessor chip of this type. The original IBM PC, the IBM PC/XT, and most of the IBM compatible PCs (PCs manufactured using IBM’s specifications) manufactured during 1984 and later, used the Intel 8088 microprocessor chip. In the 8088 chip, although the internal data path between the CPU and its associated registers is 16 bits wide, the external data path between the CPU and primary storage is only 8 bits wide, and so all operations to and from storage are done 8 bits (1 byte) at a time. Once received, however, data are processed 16 bits at a time
internally in the 8088 chip. The 8088 is thus an 8/16 bit chip. An extension of the 8088 chip is Intel's 8086 chip. Unlike the 8088, both the internal and external data paths of the 8086 chip are 16 bits wide. Ther8Q86 is4hus a true 16/16-bit chip. With an expanded built-in address bus having 20 lines, both the 8088 and the 8086 chips can identify a maximum of 1 Megabyte (220) or 1,048,576 separate storage locations. Another popular microprocessor chip is the Motorola 68000 chip, used in the Apple's Macintosh PCs. It is a 16/32-bit chip because it uses an internal data path of 32 bits and an external data path of 16 bits. So all operations to and from storage are done 16 bits (2 bytes) at a time, and data are processed 32 bits at a time internally. Thus the 68000 is more powerful than the 8088 or 8086. The address bus of 68000 chip has 24 lines, allowing a primary storage capacity of up to 16 Megabytes (224= 16,777,216 bytes). Next came the more advanced Intel chips called 80286, 80386 and 80486. The 80286 chip was about six times faster than the 8086 chip and could support a primary storage capacity of up to 16 Megabytes. It was used with the IBM Personal System/2 (PS/2) models of PCs. It was soon superceded by the more advanced Intel 80386 and 80486 chips, which were 32/32-bit chips. Other popular 32/32-bit chips were Motorola 68020 and 68030, used in some models of Apple's Macintosh PCs. Successors to the Intel's 80486 chip fall in the Pentium family of microprocessor chips - Pentium Processor, Pentium Processor with MMX Technology, Pentium II Processor, and Pentium III Processor. All these are 32/64-bit chips. That is, they use an internal data path of 32 bits and an external data path of 64 bits. So all operations to and from storage are done 64 bits (8 bytes) at a time, and data are processed 32 bits at a time internally. Thus these chips are more powerful than the 80286, 80386, 80486, 68020, or 68030 chips. Most of the current days PCs are built using the microprocessor chips of the Pentium family. In addition to wider external data path, the microprocessor chips of the Pentium family are also designed to have better clock speed so that they can execute more instructions in a given unit of time than the earlier microprocessor chips. There are two cycles that occur during the processing of each program instruction. These instruction and execution cycles are synchronized by a specific number of electric pulses produced by an electronic clock that is built into the processor. Thus, the speed with which an instruction is executed is also directly related to the chip's clock speed, or number of pulses it uses each second. Different chips (even different versions of a particular chip) are produced with different clock speeds. For example, the Intel 80286 chip used in IBM's PS/2, models 50 and 60, ran at 10 million pulses per second (that is, its clock speed was 10 megahertz or 10 MHz).
Similarly, the Motorola 68000 chip used in Apple's Macintosh system had a clock speed of 7.83 MHz Most of earlier chips had clock speed in the range of 2 to 66 MHz. The typical clock speeds of the microprocessor chips in the Pentium family are given in Figure 20.3. Note that the higher is the clock speed, the faster is the processor because it can execute more instructions in a given unit of time. Two other major techniques used by the microprocessor chip designers for better performance are larger cache sizes and larger instruction sets supported by a chip. For example, the Pentium Processor has 16 KB (16 Kilo Bytes) of cache, which is used for caching both recently used instructions and data. The more recently introduced Pentium Processor family chips (Pentium Processor with MMX technology, Pentium II Processor, and Pentium III Processor) have separate caches of 16 KB each for data and instructions. That is, they effectively have 32 KB cache size. Larger separate internal caches for data and instructions improve performance by reducing the average memory access time and providing fast access to recently used instructions and data. The instruction and data caches being separate can be accessed simultaneously. Furthermore, the Pentium II Processor and Pentium III Processor chips also have a 512 KB of level-two cache that further improves performance by allowing a large amount of recently used instructions and data to be cached. There are two kinds of memory caches. A level-one cache (known as L1 cache) is built into the CPU chip itself, and a level-two cache (Known as L2 cache) is external, typically made up of SRAM (static RAM) chips on the motherboard. A static RAM chip requires power to hold its content and has access times in the 10 to 30-nanosecond range. Static RAMs do not require refresh circuitry as do dynamic RAMs, but they take up more space and use more power. Pentium family chips Typical clock speeds Pentium Processor Pentium Processor technology Pentium II Processor
120, 133, 150, 166 and 200 MHz. with
MMX 166,200 and 233 MHz.
233, 266, 300, 333, 350, 400, and 450 MHz. Pentium III Processor 450, 500, 750 and 800 MHz. Pentium IV Processor 1 GHz and higher Figure 20.3. Typical clock speeds of microprocessor chips in the Pentium family. The Pentium Processor with MMX technology and the Pentium II Processor chips have 57 new instructions, as well as data types, as compared to the previous Intel microprocessor chips. These instructions and data types are mainly for multimedia enhancements enabling these chips to provide increased performance on a wide variety of multimedia and communications applications. Similarly,, the Pentium III Processor chip has additional new instructions enabling advanced imaging, 3D,
streaming audio and video, and speech recognition applications. Thus with a PC based on Pentium III Processor it will be possible to 1. View and manipulate higher resolution and higher-quality images than with previously available PCs, 2. Support high-quality audio, MPEG2 video and simultaneous MPEG2 encoding and decoding, and 3. Have reduced CPU utilization for speech recognition, as well as higher accuracy and faster response time. Virtually all PCs of today give users the flexibility to enhance the functionality of a basic configuration and are said to have an open architecture. For this, the system unit of a PC has expansion slots in which a wide variety of special-function add-on boards (also called add-on cards) can be plugged in. These add-on boards contain the electronic circuitry for a wide variety of computer-related functions. The number of available expansion slots varies from computer to computer. The functionality of some of the more popular add-on boards is given below. 1. Network interface card. This card, often referred to as NIC, allows a PC to be connected to a network. For example, if a PC is to be connected to an Ethernet LAN, it must be equipped with an Ethernet card. Similarly, if the PC is to be connected to an ATM network, it must be equipped with ah ATM card. 2. Fax modem card. This card allows a PC to be connected to a remote computer via a telephone-line. It also enables the PC to transfer images of hard-copy documents via telephone lines to another location just as we do using a fax (facsimile) machine. 3. Color and graphics adapter card. This card is required for interfacing a PC with video monitors that have graphics and/or color capabilities. For example, the VGA (video graphics array) board enables the interfacing of high-resolution monitors with the PC. 4. Motion video card. This card enables full-motion color video with audio to be integrated with other output on the monitor's screen. Because the card accepts multiple inputs, videos can be shown full-screen (one video input) or in windows (multiple video inputs). The system unit of a PC also has ports for connecting peripheral devices to the PC. These ports allow external input/output or storage devices (such as a printer and other devices external to the system unit box) to be plugged in to the PC in much the same way as you plug an electric equipment into an electrical socket. The two common types of ports available in almost all PCs are: 1. Serial ports. Serial ports facilitate the serial transmission of data, one bit at a time in sequence over a single wire. When a PC uses an output serial port, the 8bit bytes produced in parallel by the PC are then lined up into a single stream of bits for serial transmission to a printer or other device connected to that port. Serial ports provide an interface for low-speed printers and modems. A popular serial port standard is the RS-232C interface that facilitates the transmission of
data over telephone lines. Many serial ports installed on PC systems follow this standard. 2. Parallel ports. Parallel ports facilitate the parallel transmission of data, that is, 8 or more bits are transmitted simultaneously. A popular parallel port standard is the Centronics printer interface. Parallel ports provide the interface for such devices as high-speed printers and plotters, magnetic tape backup units, and external hard disk units. The popular operating systems for PCs are MS-DOS, MS-Windows, WindowsNT, Linux and Unix. An OS called OS/2 was also used on IBM PCs few years ago and the Apple's Macintosh PCs run the Apple's propriety OS called Macintosh OS as well as Apple's version of Unix called A/UX. Most of these operating systems enable the user to switch between tasks. This capability is known as multitasking a single-user variation of the multiprogramming concept. Multitasking eases user operation and saves lots of time when a user has to switch between two or more applications while performing a job. For example, suppose a user is using a word processor to create an annual report, and he/she needs to do some arithmetic calculations on the computer and include the calculated results in the report. Without multitasking, the user would have to close the annual report file and the word processing application, open the calculator application, make the necessary calculations, write down the results, close the calculator application, and reopen the word processing application and the annual report file. With multitasking, the user simply opens the calculator application, makes the necessary calculations, and switches back to the annual report file to continue working on it. This capability to switch between tasks (applications) saves lots of time and effort of the user. If hard disk space is sufficient, a PC can be configured to simultaneously have more than one OS. For example, a PC can be configured to have both MS-Windows and Linux OS and the user can choose at the time of switching on the PC which OS should the PC boot with depending on what application the user wants to run on the PC at that time. Thus the PC can work both as a Windows machine as well as a Linux machine. This feature provides great flexibility in the usage of a wide variety of applications for different OSs on the same PC. PCs generally cost from a few tens of thousands to about a lakh of rupees, depending on their capabilities and capacity. The tremendous growth and continuing potential of the PC market has brought many manufacturers to this arena. A few of the major PC manufacturers are IBM, Apple, Compaq, Dell, Zenith, Siemens, Toshiba, and Hewlett-Packard. WORKSTATIONS A workstation is a powerful desktop computer designed to meet the computing needs of engineers, architects, and other professionals who need greater processing power, larger storage, and better graphics display facility than what PCs provide. For example, workstations are commonly used for computer-aided design (CAD),
in which industrial designers create pictures of technical parts or assemblies. To process and view these complex and detailed diagrams, a computer with great processing power, large storage, and good graphics display monitor is needed. Workstations are also used" for simulation of complex scientific and engineering problems and for visualizing the results of simulation. They are also used for multimedia applications, such as for creating special audio-visual effects for television programmes and movies. A workstation looks very much like a PC and is typically used by only one person at a time, just like a PC. The following characteristics are often used to distinguish between the two: 1. Processing power. The processing power of a workstation is several times (typically 5 to 10 times) more than that of an average power of a PC. 2. Storage capacity. Workstations have larger main memory (typically 0.5 GB to a few GB) as compared to PCs, which have few tens or hundreds of MB of main memory. The hard disk capacity of a workstation is also much more (typically 4 GB to several tens of GB) as compared to the hard disk capacity of PCs (typically few GB). 3. Display facility. Most workstations have a large-screen (19 inch or more) color monitor capable of displaying high-resolution graphics. Thus, the color and graphics adapter card, which is optional for PCs, is available by default in workstations. PCs normally use monitors having smaller screen (17 inch or less). 4. Processor design. Unlike the microprocessor chips used in PCs for CPU, the CPU chips used in workstations are designed using a special technology known as Reduced Instruction Set Computers (RISC). The basic idea behind the RISCbased architecture is that the speed of the processor can be greatly increased by: Eliminating infrequently used programmed instructions from the set of instructions supported by a CPU chip, - Making all the instructions of equal length, and - Reducing the number of addressing modes. Popular RISC processors used in workstations are ALPHA (used in DEC-ALPHA workstations), RIOS (used in IBM workstations), SPARC (used in SUN workstations), and PA-RISC (used in HP workstations). Operating system. Unlike PCs which can run any of the five major OSs - MSDOS, MS-Windows, Windows-NT, Linux and Unix, all workstations generally run the Unix operating system or a variation of it, such as AIX (used in IBM workstations), Solaris (used in SUN workstations), and HP-UX (used in HP workstations). Also, unlike most operating systems for PCs, which are single-user oriented, a workstation's operating system is designed to enable it to support a multiuser environment. 6. Network interface card. Network H Most workstations have built-in hardware to connect to a Local Area Network (LAN). As already discussed before, this facility is normally optional in case of a
PC. Workstations generally cost from a few lakhs to few tens of lakhs of rupees depending on their capabilities and capacity. The biggest manufacturer of workstations is Sun Microsystems. Other manufacturers of workstations include IBM, DEC, Hewlett-Packard (HP), and Silicon Graphics. MAINFRAME SYSTEMS There are several organizations such as banks, insurance companies, hospitals, railways, etc. that need to process large number of transactionson-line and requirecomputer systems having massive data storage and processing capabilities. Mainframe systems are computer systems that are mainly used for handling the bulk of data and information processing of such organizations. They are also used in such environments where a large number users need to share a common computing facility such as in research groups, educational institutions, engineering firms, etc. By its very nature of usage, a mainframe system is housed in a central location with several user terminals connected to it. The user terminals act as access stations and may or may not be located in the same building in which the mainframe system is located. A typical configuration of a mainframe system consists of the following components (see Figure 20.4). 1. Host, front-end, and back-end computers. A mainframe computer system is usually composed of several computers (subordinate computers) in addition to the mainframe, or host computer. The host computer carries out most of the computations and has direct control over all the other computers. The other computers relieve the host computer of certain routine processing requirements. For example, & front-end computer is used for handling communications to and from all the user termini connected- to the computer system, thus relieving the host computer of communications-related processing requirements. Similarly, a back-end computer is used to handle data I/O operations, thus relieving the host computer of locating a particular I/O device and transferring data to or from it. Although the host computer could itself perform all these operations, the subordinate computers provide a boost to the overall system efficiency by relieving the host computer of time-consuming chores that do not require processing speed. The host computer of many modern mainframe systems are multiprocessor systems having multiple CPUs. As shown in the figure, the host and other computes are located in the systems room to which entry is restricted to system administrators and maintenance staff. 2. Console(s). One or more console terminals are also located in the systems room. These terminals are directly connected to the host processor and are mainly used by the system administrators to monitor the health of the system or to perform some system administration activities, such as changing the configuration of the system, installing new software on the system, taking system backup, etc.
3. Storage devices. For large volume data storage, a mainframe system has several magnetic d drives, which are directly connected to the back-end computer. All data to and from these magnetic disks are accessed by the host computer via the back-end computer. The magnetic disk drives are also located in the systems room. In addition, a mainframe system also has a few tape drives and a magnetic tape library for restoration or backup of data from or to magnetic tapes. The tape library is located in the systems room and is used by the system administrators to take regular backups of the data from magnetic disks on to magnetic tapes. The tape drives are located in the users room so that users who bring their input data on tape or want to take their output data on tape can have access to the tape drives. 4. User terminals. The primary means of accessing the system by the users is the user terminals, which are used by the users to write, test, and execute their programs. In the figure, although all the user terminals are shown to be located in the user room, some of them may be located at geographically distributed locations. Moreover, a user terminal need not always be a dumb terminal, which has no CPU or storage of its own and acts just as an I/O device to the mainframe system. Often a PC is used as a user terminal in which case the program development activity is carried on the user terminal and the mainframe system is used only to test and execute the program. Since mainframe systems allow multiple users to simultaneously use the system through the user terminals, their operating systems support multiprogramming with timesharing so that all the users get good response time and an illusion that their jobs are being attended to by the system. 5. Output devices. The user terminals serve the purpose of soft copy output devices. However, for hard copy outputs, «a mainframe system usually has several printers and one or more plotters connected to the back-end computer. Different printers may be of different print speeds and different print qualities. Plotters are used to produce large, high-precision hard copies of graphs, charts, blueprints, and drawings. These output devices are also located in the user room so that they are accessible to the users for taking their outputs. It is unlikely that you would find two mainframe systems configured in exactly the same way. The configuration of a mainframe system depends a lot on the type of usage and the kind of users it is meant for. The example of Figure 20.4 is just one possible configuration. Mainframe systems are much bigger and several times more expensive than workstations. A typical mainframe system looks like a row of large file cabinets and needs a large room with closely monitored humidity and temperature. A mainframe system may cost anywhere from a few tens of lakhs to a few crores of rupees depending on the configuration. A mainframe system having smaller configuration (slower host and subordinate computers, lesser storage space, and fewer user terminals) is often referred to as a minicomputer system. However,
there is no well-defined boundary for differentiating between the two. Two major vendors of mainframe systems are IBM and DEC. SUPERCOMPUTERS Supercomputers are the most powerful and most expensive computers available at a given time. They are primarily used for processing complex scientific applications that require enormous processing power. For example, scientists build models of complex processes and simulate the processes on a supercomputer. One such process is nuclear fission. As a fissionable material approaches a critical mass, scientists want to know exactly what will happen during every millisecond of a nuclear chain reaction. A supercomputer is used to model the actions and reactions of literally millions of atoms as they react. A typical scientific application like this involves the manipulation of a complex mathematical model, often requiring trillions of operations to resolve, and cannot be processed in a reasonable time without the use of a supercomputer. Some of the well known supercomputing applications (applications that need supercomputers for processing) are as follows: 1. The petroleum industry uses supercomputers to analyze large volumes of seismic data gathered during oil-seeking explorations to identify areas where there is possibility of getting petroleum products inside the earth. This helps in more detailed imaging of underground geological structures so that the expensive resources for drilling oil wells and extraction of petroleum products from them can be more effectively channelized to areas where the analysis results show better possibility of getting petroleum deposits. 2. Aerospace industry uses supercomputers to simulate airflow around an aircraft at different speeds and altitude. The advent of supercomputers has now made it possible to simulate the airflow around any complex shaped object in 3dimensional space. This has significantly enhanced the accuracy of simulation, reducing the design cycle time and cost. This in turn helps in producing an effective aerodynamic design to develop aircrafts with superior performance. 3. Automobile industry uses supercomputers to do crash simulation of the. design of an automobile before it is released for manufacturing. Doing crash simulation of an automobile on computer screen is less expensive, more revealing, and safer than crashing a real model of the automobile. This helps in producing better autonfobile designs, which are safer to ride. 4. Structural mechanics industry uses supercomputers to solve complex structure engineering problems that designers of various types of civil and mechanical structures need to deal with to ensure safety, reliability, and cost effectiveness. For example, the designer of a large bridge has to ensure that the bridge must work in various atmospheric conditions and pressures from wind, velocity, etc., and under different load conditions. Actual construction and testing of such expensive structures is prohibitive in most cases. Hence simulation techniques are
extensively used in structural mechanics area. Computer simulation of these problems is a complex activity and requires powerful supercomputers to complete the simulation in a reasonable time. Using mathematical modeling techniques, various combinations can be tried out without going for the actual construction/manufacturing of the structure and the most optimum design can be picked up. 5. Meteorological centers use supercomputers for weather forecasting. In weather forecasting and in research involving the earth's atmosphere, weather data supplied by a worldwide network of space satellites, airplanes, and ground stations are fed into supercomputers. These data are analyzed by a series of computer programs to arrive at forecasts. This analysis involves solving of complex mathematical equations, which model the atmosphere and climate processes. For example, in India, the National Center for Medium Range Weather Forecasting (NCMRWF), located in New Delhi, currently uses the CRAY XMP-216 supercomputer for predicting the medium range weather forecasts, which is very crucial for the Indian agriculture. 6. Material scientists and physicists use supercomputers to design new materials. Termed as nanostructure technology, this field requires microscopic investigation of mechanical, electrical, optical and structural properties of materials. For this, scientists need to simulate the nanoscale materials like clusters of atoms, quantum dots, quantum wires using first principles molecular dynamics technique. Such simulations are highly compute intensive since they combine the complexity of molecular dynamics and electronic structure. Hence supercomputers are needed to complete these simulations in a reasonable time. 7. Film and TV industries use supercomputers to create special effects for movies and TV programmes. Supercomputers help in producing computergenerated images using advanced graphics features in a much shorter time than the time taken to generate these images on smaller computers. The produced images are incorporated into the movies or TV programmes to create special effects. Thus the advent of supercomputers has drastically reduced the time needed to produce a full-length feature film that uses several such special effects. There are many more supercomputing applications. It is not possible to cover all of them here. All these applications are impractical, if not impossible, on mainframe systems. Supercomputers use multiprocessing and parallel processing technologies to solve complex problems faster. That is, they derive much of their speed from the use of multiple processors, and parallel processing enables a complex problem to be divided into smaller problems that can be assigned to different processors of the system and processed in parallel. A parallel program is written in such a way that the original problem is broken up into smaller computational modules, which can be allocated to different processors, and multiple processors can work
independently and cooperate to solve the problem. Thus if the original problem takes 100 hours to process on a single-processor system, and if it can be broken into 100 smaller computational modules, it can be theoretically solved in about 1 hour by using a supercomputer that has 100 processors. Since modern supercomputers employ parallel processing technology, they are also known as parallel computers or parallel processing systems. Moreover, modern supercomputers employ hundreds of processors and are also known as massively parallel processors. Parallel computers are of the following two types: 1. Shared memory parallel computer. As shown in Figure 20.5, in this type of parallel computer, there is a single system-wide primary memory that is shared by all the processors. Any communication between the processors usually takes place through the shared memory. These machines are easy to program, but the number of processors that can be usefully deployed is usually small (16 to 32 CPUs) and limited by the bandwidth of the shared memory. 2. Distributed memory parallel computer. As shown in Figure 20.6, in this type of parallel computer, the processors do not share memory, and each processor has its own local memory. All communication between the processors is done by passing messages across the network that interconnects the processors. Hence distributed memory parallel computers are also known as message passing parallel computers. They are more freely scalable and can have an almost unlimited (several hundred) number of processors. Local memory
Local memory
Local memory
Local memory
CPU
CPU
CPU
CPU
Communication network Figure 20.6. A distributed memory parallel computer. The speed of modern supercomputers is measured in gigaflops and teraflops. A gigqflop is one billion (109) floating-point arithmetic operations per. second, and a teraflop is 1012 floating-point arithmetic operations second. The PARAM 10000 supercomputer (shown in Figure 20.7) developed by the Centre for Development of Advanced Computing (C-DAC), India in the year 1998 is a 100 gigaflops system. Like mainframe systems, supercomputers also support multiprogramming and can be accessed by multiple users at the same time. However, the main difference between these two types of systems is that supercomputers primarily address
processor-bound applications whereas mainframe systems are oriented to input/output-bound applications. In processor-bound applications, the amount of work that can be performed by the computer system is limited primarily by the speed of the CPU(s), whereas in I/O-bound applications, the amount of work that can be performed by the computer system is limited primarily by the speeds of the I/O devices. Supercomputers are rarely used for processing input/output-intensive applications, such as payroll processing, accounting, or record-keeping applications. To do so would waste an expensive and relatively rare resource. Supercomputers are often treated as a national resource. This is mainly because it is not economically viable to have many such facilities for developing countries because of the cost involved in setting up the facility, maintaining the facility, and fast obsolescence of technology. Another reason for this is that a nation's leadership role in energy, space exploration, medicine, industry, and other critical areas is unlikely to continue if its scientists must use computers that are less powerful than those available to their counterparts in other countries. For example, today's supercomputers are not fast enough to simulate airflow around an entire aircraft. Builders of new airplanes must therefore simulate the passage of air around separate pieces of the airplane and then combine the results to produce an effective aerodynamic design. It is believed that the first nation to build a fast enough supercomputer, which can simulate the airflow around a complete airplane, will be the country that develops planes with superior safety, reliability, and performance. The same holds good for other scientific research and development areas. Leaders in the development of supercomputers include Cray Research Company, IBM, Silicon Graphics, Fujitsu, and Intel. In India, we have a CRAY XMP-216 supercomputer installed at the National Center for Medium Range Weather Forecasting (NCMRWF) in Delhi, and an IBM SP/2 supercomputer installed at the Supercomputer Education and Research Center (SERC) in the Indian Institute of Science, Bangalore. Realizing the importance of supercomputers in the advancement of a nation, some national efforts of building supercomputers indigenously have also been initiated. This has resulted in the development of PARAM series of supercomputers by the Centre for Development of Advanced Computing (C-DAC), Anupam series of supercomputers by the Bhabha Atomic Research Centre (BARC), and PACE series of supercomputers by the Defence Research and Development Organization (DRDO), Hyderabad. Because computer technology changes so quickly, the advanced capabilities of a supercomputer today may become the standard features next year, and next year's supercomputer will be vastly more powerful than today's. Supercomputers can typically cost from few tens of lakhs to few tens of crores of rupees depending on their processing capability and configuration.
CLIENTS AND SERVERS This is a purely role-based classification of computer systems. With the increased popularity of computer networks, it has become possible to interconnect several computers together which can communicate and interact with each other over the network. In such a computing environment, there are several resources/services, which can be shared among multiple users for cost-effective usage and can be best managed/offered centrally. A few examples of such resources/services are: 1. File server. A file server provides a central storage facility to store files of several users on the network. 2. Database server. A database server manages a centralized database and enables several users on 1 network to have shared access to the same database. 3. Print server. A print server manages one or more printers and accepts and processes print requests from any user in the network. 4. Name server. A name server translates names into network addresses enabling different computers on the network to communicate with each other. In these cases, it is usual to have one process that "owns" the resource or service and is in charge of managing it. This process accepts requests from other processes to use the resource or service. The process that owns the resource and does this management is called a server process and the computer on which the server process runs is called a server computer because it services requests for use of the resource. Other processes, which send service requests to the server, are called client processes and the computers on which the client processes run are called client computers. Note that there may be multiple client computers sending service requests to the same server computer. A generic client-server computing environment is shown in Figure 20.8. It is comprised of the following entities: 1. Client. A client is generally a single-user PC or workstation that provides a highly user-friendly interface to the end user. It runs client processes, which send service requests to the server. Generally, there are multiple clients in a clientserver computing environment and each client has its own monitor, keyboard, and mouse for allowing users to interface with the system. 2. Server. A server is generally a relatively large computer that manages a shared resource and provides a set of shared user services to the clients. It runs the server process, which services client requests for use of the resource managed by the server. A server may be a workstation, a mainframe system, or even a supercomputer. A server's operating system is generally designed to enable it to support a multiuser environment. Servers are generally headless machines, which do not have monitor, keyboard and mouse attached to them since users do not directly interact with them, but use them only via the network. 3. Network. In addition to clients and servers, the third essential component of a client-server environment is the network, which interconnects all the clients and
servers of the same computing environment. The network may be a single LAN or WAN or an internet of networks. A set of computers interconnected together to form a client-server computing environment is collectively known as a distributed computing system or distributed system. Client-server computing involves splitting an application into tasks and putting each task on the computer where it can be handled most efficiently. This usually means putting the processing for the presentation on the user's machine (the client) and the data management and storage on a server. Depending on the application and the software used, all data processing may occur on the client or be split between the client and the server. Server software accepts request for service from client software and returns the results of processing the request to the client. The client then manipulates the received results and presents it in the proper format to the user. LAN or WAN or an Internet of Networks Notice that the central feature of a client-server computing architecture is the allocation of application-level tasks between clients and servers. The computers and the operating systems of a client and a server may be different. For example, a client may be a PC running MS-Windows OS, whereas a server may be a mainframe system running Unix OS. In fact, there may be a number of different types of client computers and operating systems and a number of different types of server computers and operating systems in a single client-server computing environment. As long as a particular client and server share the same communications protocol and support the same applications, these lower-level differences are irrelevant. It is the communications protocol that enables client and server to inter-operate. One such standard communications protocol is the TCP/IP. In a client-server computing environment, it is common for one server to use the services of another server, and hence to be both a client and a server at the same time. For example, suppose a client-server computing environment has clients, a file server, and a disk block server. Any client can send a file access request to the file server. On receiving such a request, the file server checks the access rights, etc. of the user but does not actually read/write the file blocks itself. Instead, it sends a request to the disk server for accessing the request data blocks. The disk block server returns to the file server the requested data blocks. The file server then extracts the actual data from the data blocks and returns it to the client. In this scenario, the file server is both a server and a client. It is a server for the clients, but a client for the disk block server. Hence the concept of client and server computers is purely role-based and may change dynamically as the role of a computer changes. Figure 20.9 summarizes and presents a relative comparison of different types of computers based on some key features.
Points to Remember 1. Traditionally, computers were classified by their size, processing speed and cost. Based on these factors, computers were classified as microcomputers, minicomputers, mainframes and supercomputers. However, with the rapidly changing technology, this classification is no more relevant. 2. Today computers are classified based on their mode of use. According to this classification scheme, computers are classified as notebook computers, personal computers, workstations, mainframe systems, supercomputers, and clients and servers. 3. Notebook computers are portable computers which are mainly meant for use by people who need computing power wherever they go. They are small enough to fit inside a briefcase, light enough to be easily carried around, and are designed to operate with chargeable batteries so that they can be used even at places where there is no power point available. They normally run MS-DOS or WINDOWS operating system and are mostly used for word processing, spreadsheet computing, data entry, preparing presentation materials, and making presentations. 4. A personal computer is a non-portable general-purpose computer that can easily fit on a normal size office table and is generally designed to be used by one person at a time. PCs are mainly used to meet the personal computing needs of individuals either in their working places or at their homes. They are also used both by children and adults for education and entertainment. They normally run MS-DOS, MS-Windows, Windows-NT, Linux, or Unix operating system and support multitasking, which eases user operation and saves lots of time when a user has to switch between two or more applications while performing a job. 5. A workstation is a powerful desktop computer designed to meet the computing needs of engineers, architects, and other professionals who need greater processing power, larger storage, and better graphics display facility than what PCs provide. Workstations are commonly used for computer-aided design, multimedia applications, and simulation of complex scientific and engineering problems and visualizing the results of simulation. Workstations generally run the Unix operating system or a variation of it. A workstation's operating system is generally designed to support a multiuser environment. 6. Mainframe systems are computer systems that are mainly used for handling the bulk of data and information processing of such organizations as banks, insurance companies, hospitals, railways, etc. They are also used in such environments where a large number of users need to share a common computing facility such as in research groups, educational institutions, engineering firms, etc. A typical configuration of a mainframe system consists of a host computer, a front-end computer, a back-end Computer, one or more console terminals, several magnetic disk drives, a few tape drives, a magnetic tape library, several user terminals, several printers, and one or more plotters. A typical mainframe system looks like a row of large file cabinets and needs a large room. A mainframe
system having smaller configuration (slower host and subordinate computers, lesser storage space, and fewer user terminals) is often referred to as a minicomputer system. 7. Supercomputers are the most powerful and most expensive computers available at a given time. They are primarily used for processing complex scientific applications that require enormous processing power. Some of the well known supercomputing applications include analysis of large volumes of seismic data, simulation of airflow around an aircraft, crash simulation of the design of an automobile, solving complex structure engineering problems, weather forecasting, microscopic investigation of mechanical, electrical, optical and structural properties of materials, and creating special effects for movies and TV programmes. Like mainframe systems, supercomputers also support multiprogramming. However, the main difference between these two types of systems is that supercomputers primarily address processor-bound applications whereas mainframe systems are oriented to input/output-bound applications. 8. Supercomputers use multiprocessing and parallel processing technologies to solve complex problems faster, and hence they are also known as parallel computers or parallel processing systems. Moreover, modern supercomputers employ hundreds of processors and are also known as massively parallel processors. 9. Parallel computers are of two types - shared memory parallel computer and distributed memory parallel computer. In a shared memory parallel computer, there is a systemwide primary memory that is shared by all the processors and any communication between the processors usually takes place through the shared memory. On the other hand, in a distributed memory parallel computer, the processors do not share memory, and each processor has its own local memory. All communication between the processors is done by passing messages across the network that interconnects the processors. Hence distributed memory parallel computers are also known as message passing parallel computers. 10. A client-server computing environment is comprised of multiple clients, one or more servers, and a network. A client is generally a single-user PC or workstation that provides a highly user-friendly interface to the end user. It runs client processes, which send service requests to the server. A server is generally a relatively large computer that manages a shared resource and provides a set of shared user services to the clients. It runs the server process, which services client requests for use of the resource managed by the server. The network may be a single LAN or WAN or an internet of networks. 11. A set of computers interconnected together to form a client-server computing environment is collectively known as a distributed computing system or distributed system. 12. Client-server computing involves splitting an application into tasks and putting each task on the computer where it can be handled most efficiently. The computers and the operating systems of a client and a server may be different. As
long as a particular client and server share the same communications protocol and support the same applications, these lower-level differences are irrelevant. In a client-server computing environment, it is common for one server to use the services of another server, and hence be both a client and a server at the same time. Hence the concept of client and server computers is purely role-based and may change dynamically as the role of a computer changes. Questions 1. List out some of the parameters that were traditionally used to classify computers. What were the major types of computers based on this classification scheme? Why is this classification scheme no more valid? 2. On what basis are computers classified today? How many types of computers are there based on this classification scheme? 3. What is a notebook computer? Why is it so called? What are its main uses? 4. What is a laptop personal computer? Why is it so called? Describe a typical configuration of a laptop PC. 5. What is a docking station? Why is it used? 6. Notebook computers can be used even while traveling in a train or airplane. Discuss this statement. 7. What operating systems do notebook computers normally use? 8. What features make a notebook computer portable? 9. What is a personal computer? Describe its typical uses. 10. PCs have changed the work culture and work habits of numerous organizations and individuals. Discuss this statement. 11. Describe a typical configuration of a PC. 12. Which are the two commonly used models of PCs? How are they different? 13. How are a microprocessor, a motherboard, and a personal computer related? 14. Differentiate among the following types of microprocessor chips: (a) 8/8 bit chip, 8/16 bit chip, and 16/16 bit chip. (b) 16/32 bit chip, 32/32 bit chip, and 32/64 bit chip. 15. Why do some 16-bit PC systems have a maximum primary storage capacity of 1 Megabyte while others can manage up to 16 Megabytes? 16. What is meant by the "clock speed" of a microprocessor chip? How does it influence the processing capability of a processor? 17. List out three major techniques used by the microprocessor chip designers for better performance of a processor? 18. How does a larger cache size help in better performance of a processor? 19. Differentiate between LI and L2 caches. 20. What is an expansion slot? How is it useful for a buyer of a PC? 21. What is an add-on card? List out four functional enhancements that can be added to a PC by inserting one or more add-on cards into expansion slots. 22. Write short notes on:
(a) Fax modem card (b) Expansion slot (c) Add-on card (d) Serial port (e) Parallel port (f) Network interface card (g) Motion video card (h) Color and graphics adapter card 23. What is a "port"? What are the two common types of ports available in PCs? How are they different from each other? 24. What operating systems do PCs normally use? 25. What is "multitasking"? How is it different from "multiprogramming"? 26. Differentiate among multitasking, multiprogramming and multiprocessing. 27. Can a PC be configured to simultaneously have more than one operating system? If yes, explain how is this feature useful. If no, give justification for your answer. 28. What is a workstation? Wnat are its common uses? 29. List out the major differences between a PC and a workstation. 30. What is a Reduced Instruction Set Computers (RISC) technology? How does it help in increasing the speed of a processor? 31. What operating systems do workstations normally use? 32. What is a mainframe system? What are its typical uses? 33. Describe a typical configuration of a mainframe system. 34. Differentiate among the host computer, front-end computer, and back-end computer of a mainframe system. 35. What is the purpose of a console in a mainframe system? 36. What are the various types of storage devices normally used in a mainframe system? 37. What are the various types of input/output devices normally used in a mainframe system? 38. The components of a mainframe system are normally divided into two types - those that are placed ii system room and those, which are placed in user room. List out the main components of a mainframe system and mention against each component as to which of the above two types it belongs to. 39. What is a minicomputer system? 40. Discuss how the subordinate computers of a mainframe system help in improving the overall performance c the system. 41. What is a supercomputer? What are its primary uses? 42. Give examples of at least five supercomputing applications. Explain why do these applications require supercomputer. 43. Explain the two key technologies that modern supercomputers employ for solving complex problems faster.
44. What is parallel processing? Explain with an example how parallel processing can help solve a complex problem faster. 45. Why are modern supercomputers also known as massively parallel processors? 46. What is a parallel computer? How many types of parallel computers are there? Explain the differences between them and list out their relative advantages and disadvantages. 47. What is the main difference between a supercomputer and a mainframe system? 48. Supercomputers are often treated as a national resource. Discuss this statement. 49. What is a message passing parallel computer? Why is it so called? 50. Write short notes on: (a) Parallel processing (b) Parallel computers (c) I/O-bound applications (d) Massively parallel processors (e) Processor-bound applications 51. List out the main components of a client-server computing environment. Describe the roles of each component. 52. Give three examples of resources/services that can best be handled by a server in a client-server computing environment. 53. Is it possible to have a number of different types of client computers and operating systems, and a number of different types of server computers and operating systems in a single client-server computing environment? If no, explain why? If yes, explain how will different clients and servers interact with each other. 54. In a client-server computing environment, is it possible for a computer to be both a client and a server at the same time? If no, explain why? If yes, give an example of such a scenario. 55. Write short notes on: (a) Client computer (b) Server computer 56. Write short notes on: (a) Workstations (b) Mainframe systems (c) Supercomputers (c) Distributed system (d) Client-server computing environment (d) Notebook computers (e) Personal computers (f) Classification of computers 57. List out the key desirable features of an operating system for each of the following types of computers: (a) Workstations (e) Notebook computers
(b) (c) (d) 58. (a) (b) (c) (d)
Supercomputers Client computers Server computers Write short notes on: Multiprogramming Multiprocessing Multitasking Parallel Processing
(f) Personal computers (g) Mainframe systems
Glossary The purpose of this glossary is to present definitions for some of the terms that are often used in the field of computers and data processing. 2 2G technology Second-generation wireless-networking technologies. 3 3G technology Third-generation wireless-networking technologies. A A Programming Language (APL): A powerful high-level language that is well suited for specifying complex algorithms. It is a real-time language usually used in an interpretive and interactive manner and was developed primarily for scientific applications. Abacus: The earliest device that qualifies as a digital computer. It permits the user to represent numbers by the position of beads on a rack. Simple addition and subtraction can be carried out rapidly and efficiently by positioning the beads appropriately. Access control: Deals with the policies and mechanisms to prohibit a user/process from accessing those resources/information that he/she/it is not authorized to access. Access time: The time interval between the instant at which data is called for from a storage device and the instant delivery begins. Accumulator register (A): A register, which holds the initial data to be operated upon, the intermediate results, and also the final results of processing operations. Active database: A feature that makes a database capable of monitoring and reacting to specific circumstances of relevance to an application. Activity ratio: The ratio of the total number of records in transaction file and the total number of records in master file. Ada: A multi-purpose programming language developed for and used primarily by the U.S. Department of Defense in military applications. Adapter card: See add-on board. Adder: A logic circuit capable of forming the sum of two or more quantities. Add-on board: A circuit board that provides additional capabilities for a computer. Also known as adapter card expansion board, add-on card.
Address: An identification, represented in the form of < name, label, or number, for designating a particular location in storage area. Address register: A local storage register, which contain the address of the next instruction to be executed. Algorithm: A sequence of precise and unambiguous instructions for solving a problem in a finite number of operations. ALGOrithmic Language (ALGOL) A programming language that was designed for specifying algorithms, primarily of scientific nature. It was widely used in Europe. Alphabetic: Pertaining to a character set that contains the letters A to Z and the blank character.Glossary Alphanumeric: Pertaining to a character set that contains letters, digits, and usually other special characters such as the comma, dollar sign, plus sign, etc. American National Standards Institute (ANSI): A U.S. based national organization that establishes uniform standards in several fields. American Standard Code for Information Interchange (ASCII): A standard coding system for computers. ASCII-7 is a 7-bit code and its extended version," ASCII-8, is an 8-bit code. Analog cellular: A wireless communication technology in which computing equipment that are equipped with a modem can transfer data over analog cellular telephone networks. Analog computer: A computer that operates on data, which is in the form of continuously variable physical quantities such as electrical current. Anonymous ftp site: On the Internet, a computer that allows a user to log in with a username of anonymous and then a password that is the user's e-mail address. Antivirus utility: A program used to cure a computer from virus infection. Application programmer: A programmer who prepares application software. Application software: A set of one or more programs designed to solve a specific problem or do a specific task such as pay calculation, processing of examination results, stores accounting and inventory control, etc.
Archie: A keyword-based Internet search tool that allows files, stored at anonymous sites of the Internet, to be searched either by their names or by their contents. Architecture: The organization and interconnection of the various components of a computer system. Archiving: The process of storing old data files on separate disks or tapes to make room for new files on the hard disk of a computer. Arithmetic Logic Unit (ALU): The unit of a computing system, which performs all mathematical and logical operations. It is one of the components of the central processing unit (CPU) of the computer. Artificial intelligence: A branch of computer science that deals with computers that possess reasoning, learning, and thinking capabilities that resemble those of human beings. Assembler: A computer program which translates an assembly language program to its machine language equivalent. Assembly language: A programming language, which allows instructions and storage locations to be represented by letters and symbols instead of numbers. Also known as symbolic language. Assembly program: A program written in an assembly language. Also known as symbolic program. Asynchronous communication: Data transmission mode in which the sender can send data at any convenient time and the receiver will accept it. Automated tape library: A mass storage device, which uses a set of magnetic tapes and tape drives with a controller mounted in a single unit. All the tapes of an automated tape library form a single large storage unit. They are typically used for data archiving and as an online data backup device for automated backup. Automatic Teller Machine (ATM): An unattended machine used in banks that allows a customer to deposit or withdraw cash by the use of an electronic card without the need to interact with a bank employee. Auxiliary memory: A storage that supplements the main memory of a computer. Often referred to as secondary storage, this section of the computer's memory is
nonvolatile and has low cost per bit stored, but it generally has an operating speed far slower than that of the primary storage. B Backbone: The main trunk lines of a computer network that facilitates the interconnection of other computers and computer networks on the same network. Background processing: The automatic execution of lower-priority (background) computer programs when higher-priority (foreground) programs are not using the system resources. Backup: Alternate facilities of programs, data files, hardware equipment, etc. that are used in case the original one is destroyed, lost or fails to operate. Backup file: A copy of a file, created as a safety precaution against loss of data caused due to corruption or inadvertent deletion of the original file. Band printers: See chain printers. Bandwidth: The range of frequencies available for data transmission. The wider the bandwidth of a communications system, the more data it can transmit in a given period of time. Bar codes: Representation of alphanumeric data by a combination of adjacent vertical lines (called bars) by varying their width and the spacing between them. A barcode reader is a scanner used for reading (decoding) bar-coded data. Base: The total number of digits (symbols) available to represent numbers in a positional number system. Also known as radix. Basic Input Output System (BIOS): Programs stored in a ROM chip of a computer that checks the input and output devices when the system is started. Batch processing: The running o£, several computer programs one after another without the need of a human operator to run each program individually. Baud: A unit for measuring data transmission speed. It is used to describe the capacity of a carrier. In general usage, a baud is identical to bits per second. Beginners: All-Purpose Symbolic Instruction Code (BASIC) An easy-to-learn, high-level, interactive programming language that is frequently used with personal computers and in timesharing environments.
Bernoulli disk: A magnetic disk-based storage medium in which a single hard disk platter is encased in a plastic cartridge. Beta testing: The use of beta version of a software by a selected set of users for identifying errors that might have been overlooked during the regular testing phase. Beta version: The version of a software that is released to a selected set of users for testing before its actual release. Binary: A characteristic or property involving a selection, choice, or condition in which there are two possibilities. Binary Coded Decimal (BCD): One of the early coding systems, which is based on the idea of converting each digit of a decimal number into its binary equivalent rather than converting the entire decimal value into a pure binary form. For example, the decimal number 42 is represented by 0100 0010 in 8-4-2-1 BCD notation. Binary number system: A number system with a base of two. It consists of two digits: 0 and 1. Biometric device: A device used for user authentication that verifies some physical characteristic of a user, such as the person's appearance, fingerprints, hand geometry, voice, or signature. Bit: Acronym for binary digit, which stands for one binary piece of information. This can be either 0 or 1. Bitmap: A method of displaying graphic images on a monitor by altering the color (or turning on or off) of individual dots (pixels) on the coordinate system. Block: A group of related items (records, characters, etc.) handled as a unit during input and output. A section of program code treated as a unit. Blocking factor: The process of grouping two or more records together to form a block of data is known as blocking and the number of records put together in each block is called blocking factor. Boolean algebra: An algebra that deals with logical propositions, which are either true or false. This algebra is suitable for use with binary number system and is very useful in designing logic circuits used by the processors of computer systems.
Boolean function: A mathematical function in Boolean algebra. For example, w = x + y • z. Boolean variable: A variable used in Boolean algebra. It can assume a value true or false. Booting: The process of automatic loading of the operating system to primary storage and readying a computer system for use. Branch instruction: An instruction that transfers program control to a nonconsecutive instruction. Bridge: An internetworking tool that is used to connect networks that use the same communication protocols above the data-link layer but may or may not use the same protocols at the physical and data-link layers. Broadband channel: Communication channels that are capable of data transfer rates of 1 million baud (bits/second) or more. They are used to handle high volumes of data and for high-speed computer-to-computer communication or for simultaneous transmission of data to several different devices. Browser: A software tool that provides several navigation facilities to help users do Internet surfing easily and effectively. Buffer: A small storage area used to store information on a temporary basis for compensating the difference in rates of flow of data between various computer devices. For example, when data flows from an I/O device to the CPU, it passes through a buffer. Bug: An error in a computer program. Built-in-functions: Subprograms which are provided with the language so that the programmer need only call them in a program to have them automatically invoked. Also known as intrinsic subprograms. Bulletin Board System (BBS): The electronic counterpart of a wall-mounted bulletin board that allows users of a computer network to exchange information (by posting or reading messages) via a centralized database. Bundled: The inclusion of software with the hardware of a computer system as part of th'e system's total price.
Bus: A set of wires, which carries a group of bits in parallel and has an associated control scheme. It provides a communication path between two or more devices of a computer system. For example, the path between the memory and a peripheral device. Byte: A fixed number of adjacent bits that represent a particular character or symbol. Normally a byte consists of eight bits. C C: A programming language that was designed to have the virtues of high-level programming languages (mainly machine independence) with the efficiency of an assembly language. C++: A programming language that contains all the elements of the basic C language but has been expanded to include numerous object-oriented programming features. Cache memory: A small high speed memory, which is used to increase the speed of processing by making current programs and data available to the CPU at a rapid rate. Call statement: A program statement, which transfers program control to a subroutine. Canonical form: A Boolean function whose terms contain all variables (or their complements). This is the unreduced form of the Boolean function in minterms or maxterms form. Carrier: Any device that is used to transmit data from one location to another. Cathode Ray Tube (CRT) An electronic tube with a TV like screen upon which information may be displayed. C-band transmission: The use of 4 GHz band of frequencies for transmission and retransmission of microwave signals in a satellite communication system. CD-ROM jukebox: A mass storage device, which uses a set of CD-ROM disks and disk drives with a controller mounted in a single unit. All the CD-ROM disks of a CD-ROM jukebox form a single large storage unit. They are typically used for archiving read-only data. Cell: The intersection of a row and column in a spreadsheet. A user enters text, number, or formula in a cell.
Cellular Digital Packet Data (CDPD): A wireless communication technology in which digital data to be transferred is broken into small packets and inserted into analog cellular telephone networks. Central Processing Unit (CPU): The control unit and the arithmetic logic unit of a computer system are jointly known as the CPU. It is the brain of any computer system in which all calculations and comparisons performed by a computer are carried out. The CPU is also responsible for activating and controlling operations of other units of a computer system. Chain printers: Line printers that print characters by striking a set of hammers on an inked ribbon and paper placed against a metallic chain/band on which all the characters of the character set supported by the printer are embossed. Also known as band printers. Channel: (1) A path for carrying signals between a source and a destination. (2) A track on a magnetic tape or a band on a magnetic drum. Character addressable storage: A storage device in which each character has one unique location with its own address. Another name for character addressable storage is variable-word-length storage. Character printer: A printer with a print mechanism that prints one character at a time. Charge-Coupled Device (CCD): A completely electronic storage device fabricated on semiconductor chips. It stores data as pockets of charge in a semiconductor. Chip: A thin wafer of silicon on which integrated electronic components are deposited. Circuit switching: The simplest method of data communication in which a dedicated physical path is established between the sending and receiving stations through the nodes of the network for the complete duration of information exchange. Class: In OOP, a class is a description of one or more similar objects. Client computer: In a client-server computing environment, a computer that is used by an end user to interact with the server computer. It runs client processes, which send service requests to the server.
Client-server computing: A network of computers that consists of a server computer (such as, file server, database server, print server, name server) and multiple client computers that share programs and data from the server. Clip art library: A collection of graphic images stored on disk that can be readily used in word processing or desktop publishing documents, or in presentation graphics, or for other purposes. Clock speed: The number of times that the built-in electronic clock of a computer system pulses in one second. Clone: A hardware or software product that emulates an established product which has good reputation in the market. Coaxial cable: A cable made up of a single wire in the center of an insulator wrapped in mesh. It is often used as a medium to transmit data between computers and between computers and peripheral devices. Code: (1) The statements of a computer program. (2) A set of rules outlining the way in which data may be represented (like BCD, EBCDIC, ASCII). Code generator: A very high-level programming language in which programmers specify, through an interactive dialog with the system, the design specifications of a system (which tasks are to be performed) and it automatically generates (produces) the program source code for the system. Coding: The process of writing computer instructions in a programming language. Collate To combine items from two or more sequenced files into a single one. Collating sequence: An ordering assigned to the characters of a character set to be used for sequencing purposes. A commonly encountered collating sequence involves the special characters, letters, and digits, in that order. Combinational circuit: A group of logic gates interconnected to form a logic circuit. Command interpreter: The part of an operating system that serves as an interface for the user to communicate with the computer via its operating system. Command-line interface: A textual user interface in which the user gives instructions to the computer by typing commands using the keyboard.
Comment: An entry in a computer program for the purpose of documentation or explanation. Comments are used within a program to assist anyone reading the source program listing. They do not contain any program logic and are ignored (not translated) by the language processor. Commercial online service: A method of connecting computer to the Internet, in which a service organization offers a complete online environment and access to some or all of the Internet's services by using telephone service to connect a client's local call to the gateway computer o the service organization. COmmon Business Oriented Language (COBOL): A high-level programming language developed for business data processing applications. Common carriers Data transmission service providers. Communication protocol: A set of rules and procedure established to interconnect and communicate between computers. It provides a method for orderly and efficient exchange of data by establishing rules for the proper interpretation of controls and data transmitted as raw bits and bytes. Communications processor A processing unit that coordinates networks and data communications. Within a computer network, it ensures that data flows to and from different computer systems correctly and efficiently. Communications satellite: A microwave relay station precisely positioned 36000 kms. above the equator with an orbit speed that exactly matches the earth's rotation speed. It is used for data transmission between any two randomly chosen points in a very-very large area. Communications software Software that enable transfer of data and programs from one computer system to another in a network of computers. Compact Disk-Read-Only Memory (CD-ROM): A shiny, silver color metal disk of S'/i-inch (12 cm) diameter having a storage capacity of about 650 Megabytes. It is so called because of its enormous storage capacity on a compact-size disk, and because it is a read-only storage medium. Compile: To translate a program written in a high-level language into its equivalent machine language program. Usually a single source statement yields more than one machine instruction. Compiler: A translator program (software package) that translates a high-level language program into its equivalent machine language program.
Complement: For a number, which has n digits in it, a complement is defined as the difference between the number and the base raised to the nth power minus one. For example, complement of 37i0 = 102 - 1 - 37 = 6210. In Boolean algebra, the complement of a variable is the reverse (NOT) of its value. For example, complement of A is A. Completely connected network: A network in which there is a separate physical link for connecting each node to any other node in the network. Complex Instruction Set Computer (CISC) architecture: An architecture of CPUs with large instruction set, variable-length instructions, and a variety of addressing modes. Computer: An electronic device that can automatically accept and store input data, process them, and produce output results by interpreting and executing programmed instructions. Computer Aided Design (CAD): Use of computers to automate design and drafting operations. Computer Aided Instruction (CAI): A general term that refers to a learning situation in which the student interacts with (and is guided by) a computer through a course of study aimed at achieving certain instructional goals. Also called Computer Aided Education (CAE). Computer Aided Manufacturing (CAM): Use of computers to automate manufacturing operations. Computer Aided Software Engineering (CASE) tools: A set of software tools that provide a wide range of features for creation of better and more reliable software products within less time and with less effort. Computer animation: The area of computer science, which deals with the generation, sequencing, and display (at a specified rate) of a set of images (called frames) to create an effect of visual change or motion, similar to a movie film (video). Animation is commonly used in those instances where videography is not possible or animation can better illustrate the concept than video. Computer audio: The area of computer science, which deals with synthesizing, recording, and playback of audio or sound with the aid of a computer.
Computer graphics: The area of computer science, which deals with the generation, representation, manipulation and display of pictures (line drawings and images) with the aid of a computer. Computer literacy: Having sufficient computer knowledge so that an individual can work and live in a computerized society. Computer network: A distributed data processing system in which multiple computers are linked together for the purpose of data communication and resource sharing. Computer Output Microfilm (COM): An output device that records computer output on a sheet or roll of microfilm as microscopic filmed images. It uses a microfilm recorder for recording output information on a microfiche, and a microfilm reader to view the information recorded on a microfiche. Computer Supported Cooperative Working (CSCW): An application in which multiple members working on a single project and located at distant locations cooperate with each other using a network of computers. Computer system: The various components (input and output devices, storage, CPU) of a computer integrated together to perform the steps called for in the program being executed. Computer video: The area of computer science, which deals with the recording and display of a sequence of images at a reasonable speed to create an impression of movement. Computer-Based Training (CBT): Using computer technologies for training and education. The use of computers helps in making tutoring programs which are tailored to the student's level of subject mastery and can provide immediate feedback. Conference on Data SYstems Languages (CODASYL): A committee of computer users, manufacturers, and the United States government that helps to establish programming standards for various programming languages. Console In a mainframe system, the terminal that is directly connected to the host (main) computer and is mainly used by the system administrators to monitor the health of the system or to perform some system administration activities. Constant: A value, written into a program instruction that does not change during the execution of the program.
Control program: An operating system program, which controls the operations and management of resources of a computer system. The control program's major functions are job scheduling, input/output scheduling, and program monitoring. Control unit: The unit of a computer system that manages and coordinates the operations of all the other components of the computer system. It is one of the components of the central processing unit (CPU) of the computer. Cost/benefit analysis: A procedure for evaluation and selection of hardware and/or software in which lists are made of all the costs and benefits of each proposed data processing system. Counter: A device, register or storage location for storing integers that is suitably incremented or decremented to represent the number of occurrences of an event. CPU scheduler: The part of the operating system that is concerned with deciding which of the ready jobs should be allotted the CPU for execution. CPU-bound jobs: Jobs which heavily utilize the CPU operations during the course of their processing. Crash: A hardware or software failure that causes a computer system to stop functioning. Cryptography: A means of protecting private information against unauthorized access in those situations where it is difficult to ensure access control. The basic idea behind this security technique is that if it is not possible to ensure access control, it is better to prevent comprehension of information. Cursor: A highlight on the screen of a monitor that indicates the location of the next keyed-in character on the display screen. It can appear as a small rectangular object, line, arrow, or I-beam pointer, or in a spreadsheet, as a large block covering an entire cell. Cybernatics: The use of computers coupled with automatic machinery to carry out complex operations. Cyberspace: The world-wide network of computers. Cycle stealing: A memory cycle taken by the DMA controller from the CPU to enable it to transfer data directly to memory.
Cycle time: The time interval between the instant at which a read/write command is given to a memory and the instant when the next such instruction can be issued to the memory (also known as memory cycle time). Cylinder: In a disk pack, a set of corresponding tracks in all the surfaces is called a cylinder. All tracks of a cylinder are accessible by a single movement of the access mechanism. D Daisy wheel printer: A letter-quality printer that uses a printing head with the appearance of daisy and/or a wheel. Each petal of the daisy wheel has a character embossed on it. Data: A collection of facts in raw form that become information after proper organization or processing. Data Base Management System (DBMS): A set of programs provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their data in a database. Data compression: Techniques used to reduce the volume of data in such a way that the original data can be reproduced with the desired quality level whenever required. Data Definition Language (DDL): A language used to define the structure of a database. Data dependence: Dependency of data on the application. Data dictionary: A document that maintains the data definitions of a database. It is automatically created/updated by the DDL as the database schema is defined or changed. Data element: A meaningful collection of related characters. Also called a field or data item. Data entry: The conversion of human readable data into a form a computer system can interpret. Also known as data preparation. Data Flow Diagram (DFD): A graphical representation of the various inputs, processes, and outputs of a system or program.
Data integrity: Consistency of data in all files. Data Manipulation Language (DML) A language used to enter and manipulate the data in a database. Data mart: A subset of an organization's data warehouse, which contains data that is of value to a specific business unit or department rather than the whole organization. Data mining: The process of identifying interesting and useful patterns from large databases or data warehouses. It is defined as the nontrivial process of extracting implicit, previously unknown, and potentially useful information from data. Data processing: See processing Data processing system: A system that accomplishes data processing. It includes the necessary resources, which are people, procedures and devices needed to process the data. Data redundancy: Repetition of the same data items in more than one file. Data storage hierarchy: In data processing, organization of data into growing hierarchy for easy manageability. A typical data storage hierarchy consists of bits, characters, fields, records, files and database. Data transfer rate: The speed at which data is transferred from main memory to another medium on which data are recorded. For magnetic tape, the data transfer rate is equal to the product of the tape speed and recording density. In a network environment, it refers to the speed of transfer of data from one computer to another on the network. Data warehouse: A subject-oriented, integrated, time-variant, and non-volatile collection of data in support of management's decision-making process. Data warehousing: The process of creating a repository of integrated data for the purpose of decision support and analysis. Database: A collection of data files integrated and organized into a single comprehensive file system, which is arranged to minimize duplication of data and to provide convenient access to information within that system to satisfy a wide variety of user needs. Database administrator: One who is responsible for defining, updating, and controlling access to a database.
Database client: A computer system that is primarily meant for accessing data from the database in a client-server database system. It runs those modules of the DBMS that facilitate specification of access requests. Database model: The manner in which the various files of a database are linked together. The four commonly used database models are hierarchical, network, relational, and object-oriented. Database schema: The database structure definition. Database server: The computer system which stores and manages the database in a client-server database system. It runs all modules of the DBMS that deal with the maintenance and processing of data. Database software: A set of programs that enables users to create a database, maintain it (add, delete and update its records), organize its data in desired fashion, and to selectively retrieve useful information from it. De Morgan's theorem A theorem in Boolean algebra, which states how to complement a Boolean expression. Deadlock: The state of permanent blocking of a set of processes each of which is waiting for an event that only another process in the set can cause. All the processes in the set block permanently because all the processes are events that could wake up any of the other members of the set. Debugger: A software tool that helps the programmer in debugging the logic of a program by allowing him/her to follow the program's execution step-by-step and to display intermediate calculation results and field values whenever desired. Debugging: The process of finding and correcting program errors (bugs). Decimal number system: A number system with a base of 10. The ten allowable digits are 0, 1, 2, 3, 4, 5, 6, 7, 8, and 9. It is used in our day-to-day life. Decision logic: See selection logic. Decision Support System (DSS): A set of hardware and software tools to produce and present information targeted to support management decision making involving semistructured and unstructured problems. Decision table: A table used to define clearly and concisely the word statement of a problem in a tabular form. It displays the different conditions that could exist
and the different actions that the computer should take as a result of these conditions. Decoder: The part of CPU that has the necessary circuitry to decode and interpret the meaning of each and every instruction supported by the CPU. Dedicated line: See leased line. Design phase: A phase in the life-cycle of software system during which the detailed design of the system selected in the study phase occurs. Desktop Publishing (DTP): An application software package that allows a user to perform design, typesetting, and paste-up functions, and to produce high-quality, camera-ready printed pages of the documents so prepared on a high-quality printer. Development phase: A phase in the life-cycle of a software system during which a system is constructed to meet the requirements specified in the design phase. Device controller: An interface unit between an I/O device and the system bus of a computer system. On one side, it knows how to communicate with the I/O device connected to it, and on the other side, it knows how to communicate with the computer's CPU and memory through the system bus. Diagnostic routines: Programs used to print error messages by a computer to indicate system problems and improper program instructions. Dial-up line: A data transmission service that operates in a manner similar to a telephone call. Also known as switched line. Digital cellular: A wifeless communication technology in which computing equipment can transfer digital data over digital cellular telephone networks without the need for a modem. Digital computer: A computer that works with discrete quantities. It uses numbers to simulate real-time processes. Compare with analog computer. Digital library: A repository of large information in digital form. Such a library (being virtual in nature because it stores reading material only in soft-copy form) can remain open round the clock allowing users to access information from it at their own convenient time and location (from remote locations over the network).
Digital Versatile Disk (DVD): A standard format for the distribution and interchange of digital content in both computer-based and consumer electronic products. The format is designed to store various types of large multimedia data on a new and improved version of the CD which works on optical disk technology. Digitizer: An input device used for converting (digitizing) pictures, maps and drawings into digital form for storage in computers. This enables re-creation of the drawing from the stored information whenever required as well as easy incorporation of changes in the drawing as and when required. Direct file: A file organization in which the desired record (piece of information) can be directly located by its key field value without having to search through a sequence of other records. Also known as random file. Direct Memory Access (DMA) architecture: An architecture which leads to more efficient CPU utilization because the I/O request is largely processed by the DMA controller without direct participation of the CPU. Direct/leased-Iine connection: A method of connecting a computer to the Internet by using a dedicated (leased) telephone line. Directory: A mapping table used by operating systems to map file names to their corresponding file attributes and file data, and also to provide greater flexibility to users in file naming. Disaster recovery: Policies and mechanisms adopted by an organization to protect valuable data from getting lost in case its main computing facility is destroyed by a natural calamity. A copy of all valuable data is continuously created on a storage system that is geographically far apart from the main computer system. In case of a disaster, the data is recovered from this copy. Disk: A flat, circular plate coated with a magnetic material on which data can be stored by magnetization of portions of the flat surface. Disk array: A mass storage device, which uses a set of hard disks and hard disk drives with a controller mounted in a single box. All the disks of a disk array form a single large storage unit. Also known as Redundant Array of Inexpensive Disks (RAID). Disk drive: A devise used for storage and retrieval of data stored on disk. Disk pack: A removable direct-access storage medium containing multiple magnetic disks mounted vertically on a single shaft.
Diskette: A low-cost, thin, flexible magnetic disk storage device used on small computer systems. Also called a floppy disk. Distributed computing system: A computer system consisting of interconnected, multiple computers. Distributed database: Seamless integration of multiple databases located on different computers. A distributed database management system functions in such a manner that the multiple distributed databases managed by it appear as a single centralized database to the end users. Distributed operating system: An operating system for a distributed computing system in which the users view the system as a "virtual uniprocessor." Documentation: Documentation of a software system involves collecting, organizing, storing, and otherwise maintaining a complete historical record of programs and other documents used or prepared during the different phases of the system. Domain name: In a network environment, the name/address given to a computer, which uniquely identifies it. Dot-matrix printers: Character printers that form characters and all kinds of images as a pattern of dots. Dots Per Inch (DPI): A measurement of the quality for printed output. It refers to the number of dots that can be printed per linear inch, horizontally or vertically. Downloading: Moving a file from a remote computer to ones own computer in a network environment. Downtime: The period during which a computer is malfunctioning or not operating correctly due to machine failures. Drilling down: The process of viewing information from a higher level (in the hierarchy) to a more detailed view of information. Drum printers: Line printers that print characters by striking a set of hammers on an inked ribbon and paper placed against a solid cylindrical drum with characters embossed on its surface in the form of circular bands. Dual bus architecture: An interconnection architecture in which there are separate set of buses for interconnecting the CPU with memory and with I/O devices.
Dumb terminal: A terminal that has no local processing capability. Dynamic Data Exchange (DDE): A technique for linking applications that allows a user to move data from one application into another, such as from a spreadsheet into a word-processor. Changes made in one application are automatically reflected in linked applications. Dynamic RAM (DRAM): A type of RAM that uses an external circuitry to periodically "regenerate" or refresh the storage charge to retain the stored data. E Edit: To modify the form or format of the data by inserting or deleting characters where needed. Editor: A software used to interactively review and modify text materials and other program instructions. Education software: Software that allows computer systems to be used as a teaching and learning tool. Electrically EPROM (EEPROM): An EPROM chip in which the stored information is erased by using high voltage electric pulses. Also known as flash memory. Electronic card: A small plastic card having encoded data appropriate for the application for which it is used. An electronic-card reader, which is normally connected to a computer, is used to read the data encoded on an electronic card and transfer it to the computer for further processing. Electronic Data Interchange (EDI): The use of computers and data communications to transmit data electronically between companies. Electronic Data Processing (EDP): Pertaining to data processing equipment that is predominantly electronic, such as an electronic digital computer. Electronic Funds Transfer (EFT): A general term referring to a cashless approach used to pay for goods and services. Electronic signals between computers are often used to adjust the accounts of the parties involved in a transaction. Electronic mail: A service on the Internet that allows an Internet user to send a mail (message) to another Internet user in any part of the world in a near-real-time manner. Popularly known as e-mail.
Electronic pen: An input device that serves as an effective point-and-draw device. Electrostatic printer: A high-speed printer that uses charged pins to form character matrices on chemically treated paper. Electrothermal printer: A high-speed printer that uses heated elements to create characters as matrices of small dots on heat-sensitive paper. Elementary data item: A data item, which is not broken down into smaller units. E-Mail See electronic mail. Emulator: A program that permits one computer to execute the machine-language instructions of another computer of different make. Encryption: The process of transforming an intelligible information into an unintelligible form. End user: Any individual who uses the information generated by a computer based system. End-of-tape (EOT) marker: A reflective marker that indicates the end of the usable tape. End-to-end solution: A computerized solution offered to a user in which a single vendor takes the responsibility of supplying all the hardware and software components of the solution. Also known as turnkey solution. Enhanced Small Device Interface (ESDI): A drive interface standard that can address up to 1 terabyte (1012 bytes) of disk space and can reliably transfer up to 3 Megabytes per second. Entertainment software: Software that allows computer systems to be used as an entertainment tool. Erasable Programmable Read Only Memory (EPROM): A memory chip in case of which it is possible to erase information stored in it and the chip can be reprogrammed to store new information. Economically designed I/O devices: I/O devices designed in such a way that they aim at achieving maximum efficiency of the users working with them by designing them in such a way that they can be used with maximum comfort and protect the users from accidents or medical problems caused due to continuous usage of a device for long hours.
Ethernet: The most widely used LAN technology. Execution time: The total time required to execute a program on a particular system is called its execution time for that computer system. Expansion board: See add-on board. Expert system: A software package that stores data pertaining to a specialized area (known as knowledge-base) and analyzes this knowledge-base to make recommendations to users when it receives user queries. These recommendations help the users in their decision-making. Expert systems employ artificial intelligence techniques. Explicit parallelism: A technique, used in EPIC-based processors for improved performance, which allows the software to communicate explicitly to the processor when operations can be done in parallel. Explicitly Parallel Instruction Computing (EPIC): A technology, which is fast emerging as the technology for next-generation processors. The three key features used in this technology for improved processor performance are explicit parallelism, predication and speculation. Extended Binary Coded Decimal Interchange Code (EBCDIC): An 8-bit coding system, developed by IBM, that is used to represent characters in many computers. Extended Industry Standard Architecture (EIAS) bus: A 32-bit data bus that can transmit data along either 16 or 32 data lines. It was designed to accept and use the old ISA expansion slots. External fragmentation: A situation when enough total free memory space exists to satisfy the memory need of a process, but still the process cannot be loaded because the available free memory is not contiguous (is fragmented into multiple free blocks of memory). F Facsimile (FAX): Transmission of pictures, texts, maps, graphs, etc., over transmission lines, phone lines, and other carriers between geographically separated points: An image is scanned at a transmitting point and duplicated at a receiving point. Feasibility study: A study to determine whether the proposed solution is technically and economically feasible in all respect.
Field In a record a meaningful collection of one or more related characters treated as a unit. Fifth generation computers: Computers built since 1989 till now. They use ultra large-scale integrated circuits, very fast and large main memory and disk storage, very powerful programming languages and operating systems. File: A collection of related information. File Management System (FMS): A set of programs provided to facilitate the users in organizing, creating, deleting, updating, and manipulating their files. File Transfer Protocol (FTP): An application level protocol that allows a user to move a file from one computer to another in a network environment. File utilities: A set of routines, which perform a variety of generalized operations (such as, sorting, searching, merging, copying, printing and maintenance) on data files. Firewall: A software that usually runs on the proxy server of an organization and controls the flow of incoming and outgoing messages from within an organization for providing better security to the organization. Firmware: A sequence of instructions (software) that is substituted for hardware and stored in read-only memory (ROM). First generation computers: Computers built between 1942 and 1955, which used vacuum tubes and were programmed in assembly language. Few examples are ENIAC, EDVAC and EDSAC. Fixed wireless system: A wireless computing system that supports little or no mobility of the computing equipment associated with the wireless network. Fixed-head magnetic disk: A magnetic disk system that eliminates the use of an access mechanism by distributing all the read/write heads over the disk surfaces. Flash memory See Electrically EPROM (EEPROM). Flip-flop: An electronic circuit, which can be placed in one out of two stable states. Each state may be used to represent a binary digit. Floating-point numbers: Signed numbers held in a fraction-exponent format. For example, 3216 would be represented as 0.3216 x 104 in floating-point notation.
Floppy disk: See diskette. Flowchart: A pictorial representation that uses predefined symbols to describe either the logic of a computer program (program flowchart) or the data flow and processing steps of a system (system flowchart). Flowcharting: The process of drawing a flowchart for an algorithm. Font: A complete set of characters with the same style and size. Foreground processing: Automatic execution of high-priority (foreground) jobs by preempting resources held by low-priority (background) jobs. Contrast with background processing. FORmula TRANsIation (FORTRAN): A high-level, programming language used for scientific and engineering applications. Fourth generation computers: Computers built between 1975 and 1989. They used large-scale integrated circuits, semiconductor memories and powerful high-level languages and operating systems. Fourth-Generation Languages (4GLs): Programming languages that require the programmer to specify only "what" to do, not "how" to do it. Frequency Division Multiplexing (FDM): A method used to concurrently transmit data between several transmitters and receivers over a single transmission medium. The available bandwidth of a physical medium is divided into smaller, disjoint logical bandwidths and each of the component bandwidths is used as a separate communications line (channel). Front-end processor: A processor designed specifically to handle the communications processing task. Its main purpose is to off-load communications processing task from the host computer, thereby the host computer can be dedicated for applications and data processing jobs. Full adder: An adder, which adds three binary digits and outputs a result bit and a carry bit. Full duplex: A data transmission mode in which data can be transmitted between a source and a destination in both directions simultaneously. Function: A subprogram that returns a single value.
Fuzzy logic: Logic that is based on rough match of attribute values instead of exact matches. The truth values are not a definitive yes or no, on or off kind, but can lie anywhere on the 0.0 to 1.0 interval of real numbers. G Garbage-in-garbage-out (GIGO): Refers to computer errors caused due to incorrect input data or unreliable programs. Gateway: (1) A computer that is connected to the Internet by using a dedicated (leased) telephone line and is used to provide Internet connectivity to other computers via it. (2) An internetworking tool that is used to interconnect dissimilar networks that use different communication protocols. Generation: In computer talk, it is a step in technology. It provides a framework for the growth of the computer industry. Gigabytes: (GB) Memory storage equal to 1,073,741,824 (230) bytes in a computer. Gigaflop: 109 floating-point arithmetic operations per second. Google: An Internet search engine. Gopher: A text-based Internet search tool that provides hierarchical collection of information of all sorts across the Internet. Graceful degradation: The ability of a system to continue providing service proportional to the level of non-failed hardware when one or more of its hardware components fail. Graphical User Interface (GUI): An interface for computer users that provides icons (pictures) and menus (list of choices) which users can select with a mouse for telling the computer what they want it to do for them. Graphics software: Software that enables users to use a computer system for creating, editing, viewing, storing retrieving and printing designs, drawings, pictures, graphs and anything else that can be drawn in a traditional manner. Gray scales: The number of shades of a color that can be presented on a monochrome (black and white) monitor's screen or on a monochrome printer's output. Groupware: See Computer Supported Co-opeative Working (CSCW).
H Hacker: A computer enthusiast who uses the computer as a source of recreation by pushing its capabilities to its limits (performance, security, storage, etc.). Half adder: An adder, which adds two binary digits and outputs a result bit and a carry bit (if any). Half duplex: A communication system in which data can be transmitted between a source and a destination in both directions, but only in one direction at a time. Hard disk: A magnetic disk storage device that uses disks made of rigid metal (frequently aluminium). The hard disk platters come in many sizes ranging from 1 to 14-inch diameter. Hard-copy output: A computer output that is permanent in nature and can be kept in paper files or can be looked at a later time when the person is n5t using the computer. For example, output produced by printers or plotters on paper. Hardware: The physical components of a computer system such as electronic, magnetic, and mechanical devices. Hashing: The process of using an address generating function (called hashing algorithm) to map a record's key field value to its storage address. Heavy-weight process: A process of a traditional system that does not support multithreading. Hertz: One cycle per second. Hexadecimal number system: A number system with a base of 16. Its digits range from 0 to F. It is commonly used as a shortcut notation for groups of four binary digits. Hierarchical database: A database model in which the data elements are linked in the form of an inverted tree structure with the root at the top and the branches formed below. Hierarchical network: A communications network in which computers are connected in a tree-like structure. High Definition Multiple Analog Components (HD-MAC): An HDTV broadcasting standard used in Europe.
High Definition Television (HDTV): A television broadcasting standard designed to provide improved picture resolution, improved color resolution, wide screen display, and large projection as compared to conventional television broadcasting standards. High-level language: A programming language whose structure is application oriented and is independent of the structure of the computer. Each statement of such a language is translated into many machine language statements. HotBot: An Internet search engine. Hotlist: A personal list of favorite URL addresses that an Internet user is likely to frequently visit in future. A user's hotlist is automatically created and stored on his/her local web client computer by a browser. Hybrid network: A network that is a combination of two or more different network topologies. Hydra printers: Multifunctional devices that perform the functions of two or more of the following office equipment - printer, fax machine, copy machine and scanner. Hypertext: Documents that are prepared by using special linking facility for easier electronic access and manipulation. HyperText Markup Language (HTML): A powerful language used for creating hypertext documents. HyperText Transport Protocol (HTTP): An Internet protocol for interaction between the computers on the Internet. I I/O ports: Sockets used to plug-in I/O devices to a computer system. I/O-bound jobs: Jobs whose CPU utilization is very low and mostly perform I/O operations during the course of their processing. Icon: A graphical object displayed on a visual display screen that is used in place of words or phrases to allow users to select program options when selected with a mouse or other pointing device.
Impact printer: A printer which prints characters by causing hammers to strike against the paper on which information is to be printed. Incremental backup: Backing up of only those files in which changes have been made since the last backup was taken. Indexed Sequential: Access indexed sequential file. Method (ISAM): See Indexed sequential file A file organization in which an index (directory) file is used to have both sequential and random access to the records of the file. Also known as the Indexed Sequential Access Method (ISAM). Industry Standard Architecture (ISA) bus: A 16-bit data bus that can transmit data along either 8 or 16 data lines. Inference Find: An Internet meta search engine. Infopedia: A single CD-ROM disk that holds an encyclopedia, or a dictionary, or a world atlas, or a dictionary of quotations, or biographies of great people, or information about all educational institutions of a particular type in a country, etc. Information Processed data obtained as the output of data processing. It is used by people to enhance understanding and to achieve specific purposes. Infoseek: An Internet search engine. Inheritance: In OOP, inheritance is a mechanism for sharing code and behavior. It allows a programmer to reuse the behavior of a class in the definition of new classes. InkJet printers: Character printers that form characters and all kinds of images by spraying small drops of ink on to the paper. Input: Data and instructions entered into a computer for processing purposes. Input device: A device used to enter data and instructions into a computer. Examples are keyboard, mouse, trackball, joystick, scanner, digitizer, voice recognition system, and vision input system. Input/Output (I/O): Pertaining to the techniques, media, and devices used for manmachine interaction. Input/Output Control System (IOCS): A set of routines for handling the many detailed aspects of input and output operations.
Input/Output register (I/O): A register, which is used to communicate with the input/output devices. Instruction: A command or order given to a computer. It normally consists of a code to indicate the operation to be performed and address(es) in memory where the operand(s) would be found. Instruction register (I): A register, which holds the current instruction that is being executed. Instruction set: A set of machine instructions that the CPU has built-in ability to execute. Integrated Circuits (ICs): Circuits consisting of several electronic components like transistors, resistors, and capacitors grown on a single chip of silicon eliminating wired interconnection between components. Integrated Drive Electronics (IDE): A drive interface standard that provides datatransfer rates of about 1 Megabytes per second under ideal conditions. Integrated Services Digital Network (ISDN): A telephonic system that provides digital (not analog) telephone and data services. Intelligence Peripheral Interface-level 3 (IPI-3): A drive interface standard that is used in minicomputer and mainframe systems for interfacing faster, larger disks (having diameter of 8 inches or more) with the main computer system. Intelligent terminal: A terminal having extensive local processing capability. It has a built-in microcomputer that can be programmed by users. Inter Block Gap (IBG): When several records are in one block, these gaps separate the blocks of records on magnetic tape. Inter Record Gap (IRG): Blank space between two consecutive records on a tape. Interactive TV A facility that allows users to select TV programmes of their interest from remote massive video archives, view at the time they wish without leaving the comfort of their homes, and interact with the programmes via VCRlike functions, such as fast-forward and rewind. Interface: Electronic circuit used to interconnect I/O devices to a computer's CPU or memory.
Internal fragmentation: A situation when a process is allocated more memory than its actual memory requirement and the additional memory allocated to the process remains unutilized because it is neither used by the process to which it is allocated nor it can be allocated to any other process for use. Internal storage: The addressable storage in a digital computer which is directly under the control of the CPU. Internet: A network of networks of computers that links many different types of computers all over the world. Internet search enginee: An application available on the WWW that helps users locate the web sites containing useful information and references to such information. Internet Service Provider (ISP): An organization that maintains one or more gateway computers and provides Internet access facility to other users by allowing them to connect their computers to its own gateway computer(s) by using a modem. Internet surfing: The process of navigating the Internet to search for useful information. Internet telephony: Refers to the use of the public Internet for voice transmission. It enables the subscriber of an Internet service to use his/her computer to make long-distance calls. Internetworking: Interconnecting of two or more networks to form a single network. The resulting network is called an internetwork. Interpreter: A language processor that translates a statement of a high-level language and immediately executes it before translating the next source language statement. It is the most common language processor for BASIC. Intranet: A network that connects the computers and networks within an organization by using the hardware and software that is used on the Internet. Intrinsic subprograms: See built-in functions. IPV4: The current Internet protocol. IPV6: The next generation Internet protocol.
Iteration logic: A program construct used to produce loops in program logic when one or more instructions may be executed several times depending on some condition. Also known as looping logic. Ixquick: An Internet meta search engine. Java: A programming language that was designed to have machine-independent compiled code. It is primarily used for Internet-based applications. Job: See process Job Control Language (JCL): A special purpose language used to describe to a computer's operating system the resource requirements of a job fed to the computer for processing. Joint Photographic Experts Group (JPEG): A standard data compression technique for still images. Joystick: An input device that serves as an effective pointing device for applications such as video games, flight simulators, training simulators, and for controlling industrial robots. Jump instruction: An instruction or signal, which, conditionally or unconditionally, specifies the location of the next instruction and directs the computer to that instruction. A jump instruction is used to alter the normal sequence control of the computer. Justification: The alignment of text at the left margin, the right margin, or both margins. K Kernel: The central controlling part of an operating system that implements the most primitive of the system's functions. It is the only part of an operating system that a user cannot replace or modify. Key field: A unique field in a record used to distinguish one record from another. Keyboard An input device that enables data entry into a computer by pressing a set of keys (labeled buttons), which are neatly mounted on a keyboard connected to the computer system. Key-to-tape: A device used to enter data onto a magnetic tape.
Kilobytes (KB): Memory storage equal to 1024 (210) bytes in a computer. Kiosk: An unattended system that is used to store information of public interest and allows common people to access the stored information as per their requirement. Knowledge Discovery in Databases (KDD): The field which deals with the techniques and tools meant for extraction of knowledge from large datasets. Ku-band transmission: The use of 11 GHz to 14 GHz band of frequencies for transmission and retransmission of microwave signals in a satellite communication system. Label: One or more characters used to identify a statement, an instruction, or a data field in a computer program. Label record: A machine-readable record that is used to identify a data file. It is the first record of the file. Landscape mode: A mode of printing in which the printed lines are parallel to the longer edge of the paper. Language processor: A software used to convert source program instructions to object or machine language instructions. Few examples are assembler, compiler and interpreter. Large Scale Integration (LSI): Integration of over 30,000 electronic components on a single chip. Laser disk: See optical disk. Laser printers: Page printers that produce very high quality output by forming characters and all kinds of images with very tiny ink particles. They use a combination of laser-beam and electrophotographic techniques To create printed outputs. Latency: In case of disk storage, the relational delay time for the desired data (sector) to come under the read/write head positioned over that track. Maximum latency time equals the time taken by disk to rotate once. Leased line: A data transmission service in which a special conditioned telephone line directly and permanently connects two computers. A leased line can be used
for both voice and data transmissions. Also known as private line or dedicated line. Library routine: A tested routine maintained in a library of programs. Light pen: A pen shaped device that is used as an input device in computers by writing or sketching on the screen of a cathode ray tube. Light-weight process: See thread. Linker: A program that is used to properly combine all the object program files (modules) of a software, and to convert them into the software's final executable form. Linux: An open-source operating system enhanced and backed by thousands of programmers world-wide. It is a multi-tasking operating system that was originally designed to be used on personal computers. LISt Processing (LISP): A programming language suitable for manipulating nonnumeric data such as symbols and strings of text. It is used in the areas of pattern recognition, artificial intelligence and for simulation of games. Load module: The final executable form of a software produced by a linker. Local Area Network (LAN): A computer network which interconnects computers and other peripheral devices within a limited geographical area of a few kilometers. Local Multipoint Distribution System (LMDS) : A wireless communication technology for supporting point-to-multipoint applications . Logic circuit: See combinational circuit. Logic error: An error that occurs when the actual logic of a program is different from the desired logic. Such errors cause the program to produce incorrect output. Logic gate: An electronic circuit which operates on one or more input signals to produce standard output signals. For example, AND, OR and NOT gates. Logic gates are the building blocks of all the circuits in a computer. Logical operators: Symbols used to show a logical relationship between two data items. Examples in FORTRAN are .EQ. for equal and .GT. for greater than. The symbols >, < and = are also logical operators.
Logo An easy-to-learn programming language, which is Loop A sequence of instructions that is executed repeatedly until a terminal condition occurs. Looping logic: See iteration logic. Low-level languages: Programming languages that normally translate from one source instruction to one object instruction. These languages are machine dependent. Lycos: An Internet search engine. M Machine language: A low-level language that is directly understandable by the computer system. Each model of a computer has a unique machine language. Macro: A sequence of frequently used operations or keystrokes bundled together as a single command or keystroke by software so that the invocation of the single command or keystroke leads to the invocation of all the operations/keystrokes in the specified sequence. This helps speed user interaction with the system. Macro flowchart: A flowchart that outlines the main segments of a program or that shows less detail. Macro instruction: An instruction in a source language that is equivalent to a specified sequence of machine instructions. Magnetic core: Tiny rings made of magnetic material that can be polarized to represent a binary 1 or 0. Magnetic disk: See disk. Magnetic Ink Character Recognition (MICR): A technology used by the banking industry for faster processing of the large volume of cheques. This technology also ensures accuracy of data entry because most of the information is pre-printed on the cheque and is directly fed to the computer. Magnetic storage: Storage devices such as disks, drums, tapes, cores, etc. that utilize the magnetic properties of materials to store data. Magnetic tape: A secondary storage device that uses a long plastic strip coated with a magnetic material as a recording medium.
Magneto-optical disk: A storage disk, which integrates optical and magnetic disk technologies to enable rewritable storage with laser-beam technology. Mail-merge: A feature of word-processing packages that enables merging of information from a database with a letter with the addresses in a database to generate personalized letters. Main memory: See primary storage. Mainframe system: A computer system that is mainly used for handling the bulk of data and information processing of such organizations as banks, insurance companies, hospitals, railways, etc. They are also used in such environments where a large number of users need to share a common computing facility such as in research groups, educational institutions, engineering firms, etc. Management Information System (MIS): An organized collection of people, procedures, and devices used to provide the right information to the right person at the right time for proper decision-making. Manufacturer-programmed ROM: A ROM in which data is burnt in by the manufacturer of the electronic equipment in which it is used. Mass storage: Storage systems that provide access to several terabytes of stored data. They are multiple units of storage media (such as multiple disks, multiple tapes, multiple CD-ROMs) as a single secondary storage device. Massively Parallel Processing (MPP): An approach to designing supercomputers in which thousands of processors are integrated within a single computer so that the supercomputer can perform thousands of tasks simultaneously. Master file: A file containing relatively permanent data. This file is often updated by records in a transaction file. Maxterm: A Boolean quantity consisting of all terms (in its normal form or complement form) ORed together. Any combination (2n for n variables) of terms and complements is permissible, provided all are included in the term. Medium Scale Integration (MSI): Integration of up to about 100 electronic components on a single chip. Megabytes (MB): Memory storage equal to 1,048,576 (220) bytes in a computer. Memory A device or medium that can accept data, hold them, and deliver them on demand at a later time.
Memory Address Register (MAR): A register, which holds the address of the active memory location and loaded from the program control register when an instruction is read from memory. Memory Buffer Register (MBR): A register, which holds the contents of the memory word read from, or written in, memory. Memory dump: Contents of memory duplicated on another storage device or printed on a printer. Memory interleaving: A technique used to minimize memory-processor speed mismatch.-. The memory is divided into multiple equal-size modules, which can be simultaneously accessed for better performance. Memory: partition In a multiprogramming system, the area occupied by each job residing simultaneously in the main memory. Menu: A list of processing choices displayed on the screen from which a user may select. Menu bar: A menu in which the options are displayed on a horizontal bar (usually at the top of the display screen) displayed across the screen. Merging: The combining of records from two or more ordered files into a single ordered file. Message: In OOP, message serves as the mechanism to support communication between objects. Message switching: A data communication method in which the data (message) to be sent is first appended with the destination address and then sent across the network to its destination. Metadata: A catalog of information that keeps track of what is where, in a data warehouse. In short, it is data about data, which provides users and developers with a road map to the information in the data warehouse. Method: In OOP, a method of an object defines the set of operations that the object will execute when a message corresponding to the method is received by the object.
Metropolitan Area Network (MAN): Networks that share some of the characteristics of both LANs and WANs. They usually interconnect computers spread over a geographical area of about 50 kilometers. Micro Channel Architecture (MCA) bus: A 32-bit data bus that transmits data along 32 data lines. Micro flowchart: A flowchart with more details. It is also called a detailed flowchart. Microbrowser: A browser meant for use with Internet-enabled mobile devices. Microcode The very basic directions that tell the CPU how to execute an instruction; located in the control unit. Microcomputer: The smallest category of computer fabricated using a microprocessor, and other integrated circuits, namely a ROM, RAM, and I/O interface chips. Microfiche: An output device that uses combined electronic, photo-optical, and electromechanical techniques to convert digital computer output to records that can be stored as rolls of microfilm or as frames of microfilm stored on cards called microfiche. Microfilm: See microfiche. Microkernel: Technology that allows operating systems to be modeled and designed in a modular fashion. Microprocessor: An IC chip that contains all the circuits needed to perform arithmetic logic and control functions, the core activities of all computers, on a single chip. Microprogram: Firmware programs residing in read-only memory (ROM). They deal with low-level machine functions and are essentially substitutes for additional hardware. Microsecond: One-millionth of a second Millisecond: One-thousandth of a second. Minterm: A Boolean quantity consisting of all terms (in its normal form or complement form) ANDed together. Any combination (2n for n variables) of terms and complements is permissible, provided all are included in the term.
MIPS: Millions of Instructions Per Second. Mnemonic: Any kind of mental trick we use to help us remember. For example, a computer may be designed to interpret the machine code of 1111 (binary) as the subtract operation, but it is easier for a human being to remember it as SUB. Mobile computing: A wireless computing system that supports mobility of the computing equipment, which the users use to access the resources associated with the wireless network. It allows mobile users to access information from anywhere and at anytime. Also known as nomadic computing. Mobile radio networks: Wireless networks which employ narrow bands for creating mobile radio network infrastructure. Modular approach: Dividing a project into segments and smaller units in order to simplify the analysis, design, and programming efforts. Modulation: The technique by which a digital signal is converted to its analog form for transmission over an analog facility. MOdulator-DEModulator (Modem): Devices used to convert digital signals (to be communicated over an analog channel such as telephone line) to analog form at the sending end and back to digital signals at the receiving end. Monitor: A popular output device used for producing soft-copy output. It displays the output on a television like screen. Motherboard: The main circuit board of a computer system. It has the CPU chip, RAM chips, ROM chips, I/O handling chips, etc. mounted on it. Also known as system board. Motion Pictures Experts Group (MPEG) video: A standard data compression technique for digital video data. Mouse: A small, hand-held input device that serves as an effective point-and-draw device in today's computers. MPEG audio: A standard data compression technique for digital audio data. MS-DOS: Stands for Microsoft Disk Operating System. A single-user operating system that was popularly used for personal computers in the 1980s.
Multi-access bus network: A network in which a single transmission medium is shared by all nodes. Multihop wireless networks: Wireless networks designed to provide wireless communications capability in an environment where there is no wired infrastructure yet connectivity must be maintained among users who can move around arbitrarily and can at times not be in direct transmission range of each other. Multimedia: Use of more than one media (such as text, graphics, animation, audio and video) for information presentation. Multimedia computer system: A computer system that has the capability to integrate two or more types of media (text, graphics, animation, audio, and video) for the purpose of generation, storage, representation, manipulation and access of multimedia information. Multimedia database: A collection of multimedia data items, such as text, graphics, animation, audio, and video. Multimedia DBMS A set of programs to define, create, store, access, manage, and query the multimedia databases. Multimedia synchronization: Deals with mechanisms to ensure the desired temporal ordering among a set of multimedia objects in a multimedia presentation. Multiple SubNyquist Sampling Encoding (MUSE) An HDTV broadcasting standard used in Japan. Multiplexing: The method of dividing a physical channel into many logical channels so that a number of independent signals may be simultaneously transmitted on it. Multiprocessing: A term used to describe interconnected computer configurations or computers with two or more independent CPUs that have the ability to simultaneously execute several programs. Multiprocessor: A computer system consisting of two or more CPUs under a common control. Multiprogramming: Interleaved execution of two or more different and independent programs by the same computer.
Multitasking: Concurrent execution of multiple jobs (often referred to as tasks of same user) in a single-user system. Multithreading: A popular technology that is used to improve application performance through parallelism. Musical Instrument Digital Interface (MIDI): An international standard developed for digital music that determines the cabling, hardware, and communications protocol needed to connect a computer with electronic musical instruments and recording equipment. N Nanosecond: One-billionth (10'9) of a second. Narrowband channel: Communication channels that handle low volumes of data, typically from 45 to 300 baud. They are used mainly for telegraph lines and low speed terminals. National Television Systems Committee (NTSC): A television broadcasting standard used in the US, Japan, and some Asian countries. Netiquette: Deals with the rules of framing messages while interacting with other Internet users. Network: See computer network. Network database: A database model in which the data elements are organized in the form of parent-child relationships. Network Interface Card (NIC): A hardware device that allows a computer to be connected to a network, both functionally and physically. Network operating system: An operating system for a distributed computing system in which the users view the system as a collection of distinct computers. Network topology: The structure of interconnection of nodes of a computer network. Neural network: A technology used to design experimental computers that can perform human-type tasks. The circuits of these computers are patterned after the complex interconnections existing among the human brain's neurons, or nerve cells.
Newsgroup: A group of Internet/Intranet users who use the Usenet facility of the Internet to exchange their views/ideas/information on some common topic that is of interest to all the group members. Node: An end point of a branch in a network, or a common junction of two or more network branches. Nomadic computing: See mobile computing. Nonimpact printer: A printer, which performs some type of operation to the paper instead of physically striking it (as in the case of an impact printer). To print characters with nonimpact printers, the paper can be sprayed with ink, magnetized, electrically charged, .heated, placed under pressure, or struck by laser beams. Non-positional number system: A number system in which each symbol represents the same value regardless of its position in the number and the symbols are simply added to find out the value of a particular number. It is very difficult to perform arithmetic with such a number system. Non-procedural languages: Programming languages that allow the users to specify "what" the output should be without describing all the details of "how" the data should be manipulated to produce that result. For example, 4GLs. Non-volatile storage: A storage medium that retains its contents when the power is turned off or interrupted. Notebook computer: A small, portable computer mainly meant for use by people who need computing power wherever they go. Numeric: Pertaining to a character set that contains the numbers 0 to 9. Object: The primitive element of a program written in an OOP language. Object Linking and Embedding (OLE): A feature found in application software packages that allows joining of documents in different applications so that changes in either document are reflected in the other document. Object program: A fully compiled or assembled program that is ready to be loaded into the computer. It results from the translation of a source program by a language processor.
Objected-oriented software design: A technology to make programs generalized and to build software systems by combining reusable pieces of program codes called objects. Object-oriented database: A database model that captures object-oriented concepts such as class hierarchy, inheritance, and methods. Object-Oriented Programming (OOP): A programming methodology that deals with solving problems by identifying the real-world objects of the problem and the processing required of those objects, and then creating simulations of those objects, their processes, and the required communications between the objects. Object-oriented programming languages: Programming languages that encapsulate data as well as sequences of operations into abstract entities called objects, providing a stronger encapsulation mechanism than procedures. Hence objectoriented programming provides a more powerful modeling tool than procedureoriented programming. Octal number system: A number system with a base of 8. The octal digits range from 0 to 7. It is commonly used as a shortcut notation for groups of three binary digits. Off-line: A device or system not directly under the control of a computer system. Off-line data entry devices Input devices through which data is recorded on some media (such as floppies or tapes) and then entered later into the computer from the media for processing. On-line: A device or system directly connected to and under the control of a computer system. On-Line Analytical Processing (OLAP): A database architecture designed specifically for effective data analysis. On-line data entry devices: Input devices, which are connected directly to the computer on which the entered data is processed. On-Line Transaction Processing (OLTP): A database architecture designed specifically for transaction processing. Operand: The part of a machine level instruction, which tells the central processor the location .of the data to be manipulated.
Operating system: An integrated set of programs that controls the resources of a computer system and provides its users with an interface or virtual machine that is more convenient to use than the bare machine. Operation code (op code): The part of a machine level instruction which tells the central processor what has to be done. Operation phase: The life-cycle phase during which the system constructed in the development phase is used. Optical bar-code reader: An input device that is able to interpret combinations of marks (bars) that represent data. Optical Character Recognition (OCR) device: An input device (a scanner) equipped with a character recognition software. It is used for inputting text documents and store them in a form suitable for doing word processing of the documents. Optical disk: A storage medium that consists of a rotating disk, which is coated with a thin metal or some other material that is highly reflective. Laser beam technology is used for recording/reading of data on the disk. Also known as laser disks. Optical fiber: Hair-thin threads of glass or plastic used as a data transmission medium. They use light signals for data transmission. Optical Mark Reader (OMR): An input device (a scanner) that is capable of recognizing a pre-specified type of mark made by pencil or pen. Any input data that is of a choice or selection nature can be recorded for OMR input. OSI protocol: A communication protocol to interconnect geographically dispersed heterogeneous computers. This protocol has been standardized by the International Standards Organization (ISO). Output: The information generated as a result of processing by a system. Output device: A device used in a computer system to supply information and results of computation to the outside world. Examples are monitor, printer, plotter, screen image projector, and voice response system. Packed decimal format: A modified form of zoned decimal number that places two decimal digits into each byte of the field with the exception of the rightmost byte, which contains one digit and the sign of the field.
Packet switching: A data communication method in which the data (message) to be sent is first split up into packets of fixed size; each packet is then appended with source and destination addresses and some control information; and then the packets are sent across the network to their destination. At the receiver end, the packets are reassembled to recreate the message. Page: A program segment that is loaded into the main memory only when it is needed for execution. Page printer: A high-speed printer with a mechanism that appears to print an entire page at one time. Paging: A memory management scheme used*"to solve the problem of external fragmentation of memory. It does this by allowing a process's memory to be noncontiguous, thus allowing a process to be allocated physical memory wherever free memory blocks are available. Palmtop computer: A battery-operated computer with personal assistance tools. It is small-enough to fit in ones pocket and operated by keeping it on ones palm. Parallel binary adder: An adder, in which all the bits of the two binary numbers are added simultaneously. Parallel port: An I/O port that can handle a higher volume of data than a serial port because it transfers 8 or 16 bits of data at a time. It is used for connecting faster I/O devices such as printer, analog-to-digital and digital-to-analog converters, etc. Parallel processing: A technology in which a multiprocessor system is used to solve complex problems faster by breaking the problem into sub-problems and allowing different sub-problems to be processed simultaneously by different processors of the multi-processor system. Parity bit: An extra bit added to a string of bits that enables the computer to detect internal errors in the transmission of binary data. Pascal: A high-level programming language named after Blaise Pascal that facilitates the use of structured programming techniques. Password: A code by which a user gains access to a computer system. It is used for security purposes.
Patch: A modifications of a software carried out between its two released versions. A patch (modification) is created to fix a newly discovered bug. All the patches are incorporated in a new release version of the software. Peripherals: The various input/output devices and auxiliary storage units of a computer system. Personal assistance software: Software that enables users to use personal computers for storing and retrieving personal information, and planning and managing personal schedules, contacts, finances, and inventory of important items. Personal Communications Services (PCS): A wireless communication technology that is used to offer greater network complexity and lower end-to-end latency in wireless networks. Personal Computer (PC): A small and inexpensive computer used by individuals for carrying out personal jobs or for applications such as entertainment, home management, and hobbies. Personal Digital Assistant (PDA): A pocked-sized computer having facilities such as calendar, calculator, notepad, etc. Phase Alternating Line (PAL): A television broadcasting standard used in India and Western Europe (except France). Phase modulation: A form of modulation in which two binary values of digital data are represented by the shift in phase of the- carrier signal. That is, a sine wave with phase = 0° represents a digital 1 and a sine wave with phase = 180° represents a digital 0. Picosecond: One trillionth of a second. Pixel: A picture element. It is used to represent one point in a raster scan display device. PL/1 (Programming Language One): A high-level programming language designed for handling both scientific and business applications. Plotter: An output device often used by architects, engineers, city planners, and others who need to routinely generate high-precision, hard-copy graphic output of widely varying sizes.
Point-and-draw devices: Input devices that can be used to rapidly point to and select a particular graphic icon or menu item from the multiple options displayed on the screen. Many of these devices can also be very effectively used to create graphic elements on the screen, such as lines, curves and freehand shapes. Some of the commonly used point-and-draw devices are mouse, trackball, joystick, electronic pen and touch screen. Point-Of-Sale (POS) device: An I/O device capable of immediately updating sales and inventory records at a central CPU and producing a printed sales transaction receipt. Portrait mode: A mode of printing in which the printed lines are parallel to the shorter edge of the paper. Positional number system: A number system in which there are only a few symbols called digits, and these symbols represent different values depending on the position they occupy in the number. The value of each digit in such a number is determined by the digit itself, the position of the digit in the number, and the base of the number system. Power On Self Test (POST): A self diagnosis test performed when a computer is first turned on. It tests a computer's memory, its storage and peripheral devices for their proper functioning and connectivity. Predication: A technique, used in EPIC-based processors for improved performance, which reduces the number of branches and branch mispredicts. Preprocessor: A language processor that translates the statements pertaining to the extensions of a standard language into a sequence of statements of the standard language before the program can be successfully compiled by the standard language compiler. Preview: A feature of a word-processing package that permits the users to see what the document will look like (almost, if not exactly) when it is printed. Primary storage: A temporary storage area that is built into the computer hardware and in which instructions and data of a program reside mainly when the program is being executed by the CPU. Printer: An output device used to produce hard-copy output.
Privacy filters: Specially designed plastic/glass plates used with monitors to prevent others sitting/standing nearby the monitor from viewing the contents displayed on it. Private line: See leased line. Procedural languages: Programming languages that require the programmer to spell out the steps in processing procedures needed to achieve a desired result. For example, FORTRAN, COBOL, BASIC languages. Procedure: A subprogram that performs an action but returns no value. Procedure-oriented programming languages: Programming languages that use procedural abstractions that encapsulate sequences of operations into procedures (also known as functions, subroutines, or subprograms). Process: A program in execution. Also known as a job. Processing: Performing a series of actions or operations on data in order to convert them into useful information. Processor: A unit of a computer system that interprets instructions and executes them. Profiler: A performance monitoring software that helps the user in analyzing the performance of the various hardware components as well as the overall performance of a computer system. Program: A set of sequenced instructions (written in a language that can be understood by a computer) used to direct and control the operations of the computer in order to solve a problem or to perform a particular task. Program Control (PC) register: A register, which holds the address of the next instruction to be executed. Program Design Language (PDL): See pseudocode. Program library: A collection of complete programs, subroutines, and program modules that have already been developed, tested, and documented usually as a result of other programming projects.
Programmable Read-Only Memory (PROM): Similar to read only memory with the exception that these chips can be reprogrammed by using special external equipment. Programmer: One who designs, writes, tests, and maintains computer programs. PROgramming in LOGic (PROLOG): A nonprocedural programming language that deals with the use of a formal logic notation to communicate some computational process to a computer. Programming language: A language used to express algorithms in computer understandable form. Protocol: See communication protocol. Prototyping: The process of developing a small working model of a final operational system to demonstrate the concepts, look, and feel of the final system. Proxy server: A computer system that is specially configured to allow other computers of an organization to interact with the Internet only through it. Pseudo disk: See RAM disk. Pseudocode: A programming analysis tool that is used for planning program logic. It is an imitation of actual computer instructions written in a natural language such as English. Also known as Program Design Language (PDL). Public-domain software: A software that is available for free or for a very nominal charge from the bulletin boards or user-group libraries on the Internet. Also known as shareware or user-supported software. Q Quantum: See time slice. Query By Example (QBE): A software tool that enables database uses to specify search criteria by providing a front end for the query language that collects the facts about a query from the user and composes query language statements internally. Query language: A language that enables users to define their requirements for extracting the desired information from the database in the form of queries.
R Radio-router technology: An emerging wireless technology designed to make links in an IP network mobile. Radix: See base. RAM disk: A block of semiconductor RAM chips (other than the RAM chips used for the main memory) used in some computers as a simulated disk to make applications execute faster. Also known as silicon disk or pseudo disk. Note that RAM disk is not really a disk at all. Random Access Memory (RAM): A memory in which the time to retrieve stored information is independent of the address where it is stored. Random file See direct file. Random/direct-access storage device: A storage device in which access to any stored information is direct, and approximately equal access time is required for accessing any information irrespective of its location. Magnetic, optical and magneto-optical disks are typical examples of such a storage device. Rapid Application Development (RAD): A methodology for developing applications software for computers. The methodology ensures quicker development and greater user involvement than traditional methods. Raster graphics: Graphic objects composed of patterns of dots called pixels. Read-Only Memory (ROM): A non-volatile memory chip in which data are stored permanently and cannot be altered by the programmer. Real-time operating system: An operating system that is designed to handle the processing requirements of real-time systems. Real-time system: A system that must satisfy the requirement of producing the desired results before a certain deadline. Record: A collection of related items of data treated as a unit. Record length: A measure of the size of a record, usually specified in units such as characters.
Reduced Instruction Set Computer (RISC) architecture: An architecture of CPUs with a small instruction set, fixed-length instructions, and reduced references to memory to retrieve operands. Redundant Array of Inexpensive Disks (RAID): See disk array. Refresh rate: The number of times the electron beam of a display monitor scans the entire screen per second. Registers: Small, high-speed storage units, which are used to hold the information on a temporary basis when there is an interpretation and execution of instructions by the CPU. Data stored in a specific register have a special meaning to the CPU. Relational database: A database model in which the data elements are organized in the form of multiple tables and the data in one table is related to data in another table through the use of a common field. Reliability: Refers to the degree of tolerance of a system against errors and component failures in the system. Remote access: Accessing a computer from a distant station using communication facilities. Remote dial-up connection: A method of connecting a computer to the Internet via a gateway computer of an Internet service provider. Report file: A file that holds a copy of a report generated by a data processing application in computer-accessible form until it is convenient to print it. Report generator: A software that enables the users of a database to design the layout of a report so that it can be presented in the desired format. Report Program Generator (RPG): A business oriented, general purpose programming language designed to generate the output reports resulting from the processing of common business applications. Response time: The elapsed time between submission of a job to the system for processing and the first response produced by the system for the job. Ring network: A computer network in which there is no host computer for controlling other computers and each computer in the network has communicating subordinates.
Rolling up: The process of viewing information from the detailed level (in the hierarchy) to a higher level view of information. Router: An internetworking tool that is used to interconnect those networks that use the same high-level protocols above the network layer. Routing: The selection of actual path to be used to transmit a message in a wide area network environment. RS-232-C interface: A standard interface used for interconnecting user terminals to computers. It was defined by the Electronics Industrial Association of USA. Run time: The time required to complete a single, continuous, execution of an object program. S Scanner: An input device used for direct data entry into the computer system from source documents. Schema builder: A software tool that enables a user to define a database schema interactively by prompting the user to enter the field name, field type, field size, and so on for each field. Screen image projector: An output device that is used to project information from a computer on to a large screen (such as a cloth screen or a wall) so that it can be simultaneously viewed by a large group of people. Script: A program written in a scripting language. Scripting language: A programming language that allows programmers to integrate several command statements (used to instruct a computer to do some job) into a single command so that the programmer need not type all the commands every time he/she needs to perform the intended task. Scrolling: Using program-specified ways (by the use of keys or mouse) to move the items displayed in a window up or down, or left or right, in order to see parts of the file that extend past the bottom or top or sides of the window. Second generation computers: Computers built during the period 1955 and 1964. They used transistors in CPU and magnetic cores for main memory. On software front, they used high-level languages like FORTRAN and COBOL for programming and batch operating systems.
Secondary storage: See auxiliary memory. Sector: The smallest unit with which any disk can work. Disk drives are designed to access only whole sectors at a time. Security: Policies and mechanisms that deal with protecting the various resources and information of a computer system against destruction and unauthorized access. Seek time: In a disk system, the time required for a read/write head to move to the track where the data to be read or written is stored. Maximum seek time equals the time taken for the head to move from the outermost track to the innermost track. Selection logic: program construct used for selecting the proper path out of two or more alternative paths in the program logic. Also known as decision logic. Self-documenting languages: Programming languages that do not require much explanation in order to be understood by someone reading the program instructions. For example, COBOL language. Semiconductor storage: A storage device whose storage elements are formed as solid-state electronic components on an integrated circuit chip. Sequence logic: A program construct used for performing instructions one after another in sequence. Sequential Couleur Avec Memoire (SECAM): A television broadcasting standard used in France, Eastern Europe, Russia, and Middle East. Sequential file: A file in which records are stored one after another in an ascending or descending order determined by the value of the key field of the records. Sequential processing: A technique in which a number of similar items or transactions to be processed are grouped together and processed in a designated sequence. Sequential-access storage device: A storage device in which information can only be retrieved in the same sequence in which it is stored. Magnetic tape is a typical example of such storage device. Serial port: An I/O port used to connect any device that can send or receive a stream of bytes serially, one bit at a time. It is used for connecting slow I/O devices such as terminal, keyboard, mouse, modem, etc.
Server computer: In a client-server computing environment, a computer that manages a shared resource and provides a set of shared user services to the clients. It runs the server process, which services client requests for use of the resource managed by the server. Shareware: See public-domain software. Silicon disk See RAM disk. Simplex: A data transmission mode in which data can flow only in one direction. Simulation: To represent and analyze properties or behavior of a physical or hypothetical system by the behavior of a system model. Single In-line Memory Module (SIMM): A memory board with memory chips on it. Memory chips can be added on the board to increase the computer's main memory capacity. SLIP/PPP connection: SLIP (Serial Line Internet Protocol) and PPP (Point-toPoint Protocol) are software that allow a client computer to be connected via a modem to a gateway computer of an Internet service provider in such a manner that all the software tools that are on the client computer can be used by a user of the client computer to access the Internet, rather than depending on the gateway computer's software. Small Computer System Interface (SCSI): A drive interface standard that uses a generic device controller (called SCSI controller) on the computer system and allows any device with an SCSI interface to be directly connected to the SCSI bus of the SCSI controller. Small Scale Integration (SSI): Integration of only about ten to twenty electronic components on a single chip. Smart card: An electronic card having a built-in microprocessor chip. Data to be encoded on the card is permanently stored in the memory of the microprocessor chip. Smart terminal: A terminal having local data editing capability and the ability to consolidate input data prior to sending them to the main computer. It has a microprocessor and some internal storage in addition to I/O capabilities. Snowflake schema: A data warehouse model, which consists of normalized dimension tables, by attribute level with each smaller dimension table pointing to an appropriate aggregated fact table.
Soft-copy output: A computer output that is temporary in nature and vanishes after use. For example, output displayed on a terminal screen or spoken out by a voice response system. Software: A set of computer programs, procedures, and associated documents (flowcharts, manuals, etc.) related to the effective operation of a computer system. Software engineering: A branch of engineering that deals with the systematic application of the principles of computer science and mathematics in creating and building cost-effective software solutions. Software package: A group of programs that solve a specific problem or perform a specific type of job. For example, a word-processing package. Software piracy: The illegal copying or use of software. Solid state: Electronic components whose operation depends on the control of electric or magnetic phenomena in solids, such as transistors and diodes. Sort: The process of arranging data into a desired sequence. Source program: A program written in a symbolic or high-level language such as assembly language, COBOL, BASIC, etc. Special character: A graphic character that is neither a letter, a digit, nor a space character; for example, the dollar sign, comma, period, etc. Special purpose computer: A computer that is meant for doing only a particular type of jobs. Speculation: A technique, used in EPIC-based processors for improved performance, which reduces the effect of memory-to-processor speed mismatch. Speech synthesizer: A voice response system that converts text information into spoken sentences. Spooling: A mechanism to reduce the speed mismatch between slow I/O devices and CPU, and to convert mutually exclusive I/O devices into non-mutually exclusive I/O devices. Spreadsheet: A numeric data analysis tool (software package) that allows users to create a kind of computerized ledger.
SQL: A standard query language for relational databases. Stack: A memory in which information, which is stored last is on top and is retrieved first. Also known as LIFO (Last-In-First-Out) storage. Stand-alone: In a distributed data processing system, it refers to a computer system that has an independent (from a central-site computer) processing and storage capability. Star network: A network in which multiple computers are connected to a central host computer. These computers are not linked directly to each other and can communicate only via the host computer. Star schema: A data warehouse model, which consists of one fact (major) table and many dimension (minor) tables. The fact table stores the primary keys for all dimension tables. Statement: In programming, an expression or generalized instruction in a source language. Static RAM (SRAM): A type of RAM that does not need any special generator circuits to retain the stored data. It is faster, costlier, and consumes more power than dynamic RAM. Storage: See memory. Storage hierarchy: A hierarchy of storage technologies (such as cache memory, main memory, secondary storage and mass storage) used in a computer system. This is usually done because a single type of storage in not superior in speed of access, capacity, and cost. Stored program computer: A computer where the program to solve a problem and the necessary data are stored in its memory. StriNg Oriented symBOlic Language (SNOBOL): A programming language designed to manipulate strings of characters and used for non-numeric applications. Structured design: A system design approach in which a difficult problem is broken into smaller problems that is small enough to manage but independent enough to solve separately.
Structured programming: An organized approach to programming involving the use of the three basic control structures - sequence, selection, and iteration, and the use of top-down concepts to decompose main functions into lower-level components for modular coding purposes. Study phase: A system life-cycle phase during which a problem is identified, alternative solutions are studied, and the most feasible solution is selected. Subprocedure: See subroutine. Subprogram: See subroutine. Subroutine: A standardized program written in such a way that it can be used as part of another program whenever necessary. A subroutine is normally invoked through other programs by the use of CALL statements. Also known as subprogram or subprocedure. Supercomputer: Computer systems characterized by their very high processing speeds. They are generally used for complex scientific applications. Superscalar architecture: An architecture which permits more than one instruction to be executed in each clock cycle. Swapping: Storing programs on disk and then transferring these programs into main storage as and when they are needed. The technique is used to process large programs or several programs with limited memory. Switched line: See dial-up line. Symbolic language: See assembly language. Symbolic program: See assembly program. Synchronous communication: Data transmission mode in which the sender and receiver must synchronize with each other to get ready for data transmission before it takes place. Syntax: A set of rules of a programming language that define the pattern or structure of the word order and punctuation of an instruction. It is analogous to rules of grammar in English language.
Syntax errors: Errors in computer programs that typically involve incorrect punctuation, incorrect word sequence, undefined terms, or misuse of terms. These errors are automatically detected and pointed out by language processors. System: A group of integrated parts (people, methods, machines, and materials) that have a common purpose of achieving some objective(s). System analysis: A detailed step-by-step investigation of related procedures to see what must be done and the best way of doing it. System analyst: The individual responsible for planning a computer data processing system. He/she utilizes tools such as flowcharts, decision tables, program design language, etc. These plans are then passed to the computer programmer. System commands: Commands used to communicate with the operating system of the computer. System programmer: A programmer who prepares system software. System software: A set of one or more programs designed to control the operation and extend the processing capability of a computer system Tape density: The amount of data that can be placed over a given length of tape. The density is usually expressed in bytes or characters per inch. Tape drive: A device used for storage and retrieval of data stored on magnetic tap medium. Telecommunications: Transmission of data between computer systems and/or terminals at different locations through telephone facilities. Telnet: An application level protocol that allows a user to log in to another computer from his/her current computer in a network environment. Template: A standard format used to quickly create new •documents in the same format. Terabytes (TB): Memory storage equal to about one trillion (1012) bytes. Teraflop: 1012 floating-point arithmetic operations per second. Terminal: See Video Display Terminal (VDT).
Terminal emulator: A program that runs on a PC or workstation and pretends that it is a terminal when the PC or workstation is connected to a network and communicates with other computers on the network. Testing: The process of making sure that the program performs the intended task. Thermal printer: A printing device that utilizes paper that is sensitive to heat. Third generation computers: Computers built between 1964 and 1975 that used integrated circuits in CPU, high speed magnetic core main memories, powerful high-level languages and saw the advent of time-sharing operating system. Thread: In a multithreading system, the basic unit of CPU utilization is a thread. A process consists of an address space and one or more threads of control. All the threads of a process share the same address space, but each thread has its own program counter, its own register states, and its own stack. Also known as lightweight process. Throughput: The total amount of useful processing carried out by a computer system within a given time period. It is measured as the number of processes that are completed by the system per unit time. Time division multiplexing: A method of sharing a communication channel in which the total time available in the channel is divided between several users and each user of the channel is allotted a time slice during which he/she may transmit a message. The channel capacity is fully utilized by interleaving a number of data streams belonging to different users into one data stream. Time log: A log documentation automatically maintained by many computer systems that describe in detail how the computer system was used during the day. Time slice: The short period of time during which a user process gets the attention of the CPU in a time-sharing system. Also known as time slot or quantum. Time slot: See time slice. Time-sharing: A mechanism to provide simultaneous interactive use of a computer system by many users in such a way that each user is given the impression that he/she has his/her own computer. It uses multiprogramming with a special CPU scheduling algorithm to achieve this.
Top-down approach: A disciplined approach to system design or program design in which top-level functions are decomposed into a hierarchy of understandable lower-level modules for better management and easy handling. Touch screen: A simple, intuitive, and easy to learn input device that enables the users to choose from available options by simply touching with their finger the desired icon or menu item displayed on the computer's screen. Track: In case of magnetic disk storage, one of many circular concentric rings used for storing data. In case of magnetic tape storage, a horizontal strip, or channel, running the full length of the tape and used for recording data. Trackball: An input device that serves as an effective pointing device. Transaction file: A file in which current data are stored for subsequent processing usually in combination with a master file. Transducer: A device capable of changing signals from analog form to digital form or vice-versa. Transistor: A controlled electronic switch fabricated using a semiconductor. It is extensively used in the design of various electronic equipment. Transponder: A device mounted on a communication satellite, which receives, amplifies and retransmits signals from earth stations. Time division multiplexing: A method of sharing a communication channel in which the total time available Trojan horse: A computer virus disguised as a useful program, but containing hidden instructions to perform some malicious task instead. Truth table: A table which gives the output values for the various input combinations in case of a logical expression. Tuple: A group of related fields (a row of a table) in a relational database. Turnaround time: The elapsed time between the submission of a job to a computer system for processing and the completion of the job. Turnkey solution: See end-to-end solution. Twisted-pair cable: A simple, inexpensive, and slow wired transmission medium.
UART: A chip on the device controller's card that converts parallel data from the system bus of the computer into serial data that can flow through a serial cable or telephone wire. Ultra Large Scale Integration (ULSI): Integration of about ten million electronic components on to a single chip. Ultra Voilet EPROM (UVEPROM): An EPROM chip in which the stored information is erased by exposing the chip for some time to ultraviolet light. Unbundled: A term, which means that the application software is priced separately from the computer hardware. Unconditional transfer: A program instruction that causes the program control to flow out of normal sequence unconditionally. Unibus architecture: An interconnection architecture in which there is a single data bus, a single address bus, and a single control bus used to interconnect the CPU with both the memory and I/O devices. Unicode: A code for internal representation of characters. It uses two bytes to represent more than 65,000 characters or symbols. It is targeted to cover all languages of the world and mark computer programs and data internationally interchangeable. Uniform Resource Locator (URL): An addressing scheme used by WWW browsers to locate sites on the Internet. Uninterrupted Power Supply (UPS): A battery-supported power unit between an external power source and a computer system that supplies clean and continuous power even during power failures. Universal gate: A logic gate, which is alone sufficient to implement any Boolean function. For example, NAND and NOR gates. Universal Product Code (UPC): A universal or standardized optical bar code which normally appears on retail packages and which is read by a laser beam scanner. Unix: A multi-user, time-sharing operating system. It is a very popular operating system and can be used on a wide variety of computers ranging from notebook computers to super computers.
Uploading: Moving a file from ones own computer to a remote computer in a network environment. Uptime: The time duration when the computer system is in operation. Upward compatible: The ability of hardware or software to function with all of the same equipment and software with which its predecessors could function. Usenet: A service on the Internet that allows a group of Internet users to exchange their views/ideas/information on some common topic that is of interest to all the members belonging to the group. User authentication: Deals with the problem of verifying the identity of a user (person or program) before permitting access to the requested resource. User-friendly: A program or computer system designed in such a manner that even individuals who lack computer experience can easily use the system without much training. User-programmed: ROM A PROM in which the user can load and store "readonly" programs and data. User-supported software: See public-domain software. Utilities: A set of programs that help users in system maintenance tasks and in performing tasks of routine nature. Value Added Network (VAN): Enhanced data transmission service provided by doing value addition to the standard services of the common carriers. Added value may include features such as electronic mail, data encryption/decryption, access to commercial databases, and code conversion for communication between incompatible computers. Variable name: In a program, the name assigned to a data field that can assume any field that can assume any of a given set of values. Vector graphics: Graphic objects composed of patterns of lines, points, circles, arcs, and other geometric shapes (vectors) which can be easily represented by few geometric parameters. Very Large Scale Integration (VLSI): An electronic circuit with about 10,000 transistors fabricated in a single silicon chip.
Very Small Aperture Terminals (VSATs): Smaller dish antennas having lower power transmitters and receivers which can be installed at users' sites, it enables a large number of users to benefit from satellite communication. Video conferencing: A system in which persons sitting at CRT screens see and talk to each other via a computer-communications network. It saves traveling cost and the valuable time of executives. Video Display Terminal (VDT): A monitor and a keyboard used together as both an input and an output device. Also known as terminal. Video Graphics Array (VGA): An add-on board that enables the interfacing of very high-resolution monitors to a computer system. Virtual classroom: A learning environment created by using computing technologies. Students can join and leave the learning environment at their own convenient time and can learn at their own pace of learning. Virtual machine: The concept of adding the operating system layer on top of the bare hardware to manage all parts of the system, and present the users with an interface that is easier to program and use. Virtual memory: A memory management scheme that allows the execution of processes that might not be completely loaded in the main memory. The three basic concepts used for the realization of virtual memory are on-line secondary storage, swapping and demand paging. Virtual reality: A technology using which the user can put a pair of goggles and glove and tour a three-dimensional world that exists only in the computer, but appears realistic to the user. Virtual shop: A shopping environment created by using computing technologies. It remains open 24 hours and allows the customers to do purchases at the time they wish without leaving the comfort of their homes. Virus: A computer virus is a piece of code attached to a legitimate program that, when executed, infects other programs in the system by replicating and attaching itself to them. In addition to this replicating effect, a virus normally does some other damage to the system, such as corrupting/erasing files. Vision input system: An input device that allows a computer to accept input just by seeing an object. The input data in this case is normally an object's shape and features in the form of an image.
Visualization: Deals with exploring data and information graphically for better comprehension. Voice recognition device: An input device that allows a person to input data to a computer system by speaking to it. Voice reproduction system: A voice response system that produces audio output by selecting appropriate audio output from a set of pre-recorded audio responses. Voice response system: An output device that enables a compute to talk to its users. Voiceband: Data communications system that handles moderate volumes of data, typically from 300 to 9600 bauds. Phone lines that we use to talk to other people is an example . Volatile storage: A storage medium that loses its contents in the event of power failure or when power is switched off. Web client: Any computer on the Internet that can access web servers. Web page: A hypertext document on the Internet, Web server: Any computer on the Internet that uses the HTTP protocol. Wide Area Network (WAN): A computer network which interconnects computers spread over a large geographical area. It may also enable LANs to communicate with each other. This type of network may be developed to operate nationwide or worldwide and the transmission medium used are normally public systems such as telephone lines, microwave and satellite links. Wildcard (character): A special character, usually a ? or a *, that is used in software commands as a generic reference to any character or any combination of characters, respectively. Winchester disk: Medium-sized, non-interchangeable metal disks permanently housed in sealed, contamination-free containers. Read/write heads are built-in with the disk. Window: A rectangular portion of a computer screen that is dedicated to specific activity or application at a particular instance of time. The display screen can be separated into several windows to simultaneously show the status of the different applications at any- instance of time.
Windows: An operating system developed by Microsoft for use on personal computers. Its various release versions are Windows 3.0, Windows 95, Windows 98, Windows 2000 and Windows XP. Windows NT: A multi-user, timesharing operating system developed by Microsoft that has Unix-like features. It is designed to be used for powerful workstations and network and database servers. Wireless Application Protocol (WAP): A technology that allows development of wireless applications that are independent of the underlying wireless access technology and that also enables access of information from existing web sites using mobile devices. Wireless computing system: A computer network that uses wireless communication technologies. It enhances the functionality of computing equipment by freeing communication from the location constraints of the wired computing systems. Wireless LAN (WLAN): A LAN having a wireless interface to enable wireless communication among the equipment that are part of the LAN. Wireless Local Loops (WLLs): A wireless communication technology designed to support fixed wireless systems. Wireless Markup Language (WML) : A version of HTML optimized for mobile devices. Word: A group of bits or characters considered as an entity and capableof being stored in one storage location. Also fixed-size storage areas that form the primary memory of a computer system. Word length: A measure of the size of a word, usually specified in units such as characters or bits. Each location of a computer system can store a fixed number of characters or bits called its word length. Word processing: The use of computers to create, view, edit, format, store, retrieve, and print text materials. Word wrap: A word processing feature that automatically moves words down to the beginning of the next line if they extend beyond the right margin. Word-addressable storage: Storage designed to store a fixed number of characters (equal to its word-length in bytes). For such a storage device, storage space is
always allocated in multiples of word-length. Its another name is fixed-wordlength storage. Workstation: A powerful desktop computer designed to meet the computing needs of engineers, architects, and other professionals who need greater processing power, larger storage, and better graphics display facility than what PCs provide. World Wide Web (WWW or W3): Network of computers across the globe (world) interconnected together on the Internet, and using the concept of hypertext to link Internet sites and information on the Internet. Worm: A computer virus designed to erase/alter the data/program from a computer system's memory/disk. Write-Once, Read-Many (WORM) disk: An optical disk that looks like a standard CD-ROM disk, but allows the user to record his/her own information on it by using a CD-Recordable (CD-R) drive. WYSIWYG: Stands for What You See Is What You Get. This feature of a wordprocessing package allows the users to display a document on the screen in a form that closely resembles what the printed document will look like. X terminal: A graphics terminal that has a rich set of commands to display characters, move the cursor position, draw graphics and images, and manage multiple windows on the screen. Yahoo: An Internet search engine. Zip disk: See Bernoulli disk. Zoned decimal number: Any numeric character coded in the Extended Binary Coded Decimal Interchange Code (EBCDIC). Each decimal digit occupies one byte of storage.
*********************************************************************** *
