REAL TIME SYSTEMS UNIT 1: INTROD INTRODUCT UCTION ION TO REAL-TI REAL-TIME ME SYSTEM SYSTEMS: S: Historical Historical background, background,
RTS Definition, Classification of Real-time Systems, Time constraints, Classification of Programs.
6 Hours
Historical background background
•
Brown and Campbell – 1950 – Earliest proposal - Computer operating in real time (as a part of the control system) Consists of feedback and feed-forward loops Assumption was that, the analog computers can be used Digital computer elements did not were excluded
•
First digital computer – developed for real time con trol – for airborne operation Digitrac digital computer – 1954 – used u sed to provide an automatic flight and weapons control system
•
Industrial control – in digital computers usage – late 1950s Computer and electronic system manufacturers – for extending markets – been out from military uses of – initiated the computer control industries 1958, September – Louisiana Power and Light Company - Daystrom computer system’s system’s installation – for plant monitoring – in their power station – at Sterling, Louisiana It was not the control system
•
First industrial computer installation Texaco Company Compan y –Ramo-Wooldridge Company - RW-300 RW-300 system’s installation at Port Arthur refinery in Texas 1959, March 15th – Refinery were using – closed loop control system (Anon, 1959)
•
1957-58 - Monsanto Chemical Company – with the cooperation of RamoWooldridge Company – studied the control by computer
•
1958, October – it planned to have computer control for ammonia plant – at Luling, Louisiana
•
1960, July 20th – Commissioning of computer control for ammonia plant at Luling, Louisiana – began
•
1960, April 4th – Closed loop control was achieved ach ieved – after rewriting – the control of the program - Noise problems – were faced – on the measurement me asurement of signals
•
1959-60, B. F. Goodrich Company – Acrylanite plant, at Calvert city, Kentucky – had the same installation scheme – as above – and also –40 systems of RW-300 RW-300 based – were in for supervisory control systems for using using – for steady-state optimization calculations – for determining the set-points – for standard analog controllers – Here, computer – not controlling con trolling directly – movement of the valves or other plant actuators
•
1959 – Plan to had the following installation were begun (Burkitt,1965)
•
1962, November – Ferranti Argus 200 systems - First DDC (Direct Digital Control) system – installed – at ICI ammonia-soda plant at F leetwood, Lancashire, UK It was the large system – 120 control loops (94 of were used actually) – 256 measurements (224 of were used actually) – in Fleetwood system
•
1961 – Monsanto Company – Texas city plant – and, a hierarchical control scheme for petrochemical companies, at Chocolate Bayou – DDC projects – began
•
RW-300 RW-300 computer – used the rotating drum store – to hold the control con trol program
•
Ferranti Argus Argus 200 – Used the ferrite core store – to hold the control con trol program PROM – program was held in it Loaded by – physically inserting pegs into a plug board – each peg representing – one bit in the memory word Was laborious to set up initially Was very reliable –since, destruction of the memory contents - can be done only by physical dislodgement of the pegs Security was enhanced – using special power supplies – and, switch-over mechanisms – to protect information held in the main core store
The information – was as follows – 1. Set points – Loss most undesirable 2. Valve Valve demand – Presence after controlled stoppage allows computer to gain control of plant immediately and without disturbing the plant (referred to as bumpless transfer) 3. Memory calculations – Loss is tolerable, soon will be updated and only slight disturbance to plant 4. Future development – Extension to allow for optimization may require information to be maintained for long periods of time 5. More rapid memory access – were in Ferranti Argus 200 – that of RW-300 RW-300 and similar machines Began the second phase of application of computers – to real time control
•
1960s, Early – Computers ware using – combined magnetic core memories and drum stores – drum eventually giving way wa y to hard disk drives
Ex.: for early 1960s computers –
General Electric 4000 series
BM 1800
CDC 1700
Foxboro FOX 1, and Foxboro 1A
SDS
Xerox SIGMA series
•
Ferranti Argus series
Elliot Automation 900 series
Cost of earlier computer increased – in attempts atte mpts to resolve some problems – and only one computer – for supervising control and DDC – can be justified only – it was with further problems in development of softwares
•
Softwares – written by specialists programmers – in machine code – it was manageable earlier – since, tasks were defined clearly and the programs’ length were less
•
Combining of DDC and supervisory control – increased – 1. the code length, for a given application 2. the complexity of the programming progra mming
DDC and supervisory control – were with w ith very different time-scales DDC control programs – have to interrupt the supervisory control programs’ increase in code length – made all the code could be stored in core memory and also swapping of code between the drum memory and core – to be to done were also - Solutions to increased code length problems above – were by Development of general purpose real-time operating systems and high-level languages
•
Late 1960s – RTOS were developed – PROCESS FORTRAN compilers had their appearance
•
Problems and costs of involving in having one computer only for use – made the users to retreat – the smaller systems – for which newly developing microcomputers (like DEC PDP-8, PDP-11, Data General Nova, Honeywell 316, etc.,) were ideally suited
•
Microcomputers less cost – made them suitable – to load the large number of tasks onto one machine, (by (b y using more than one microcomputer, instead a single computer)
•
1970 – two computers on the systems were using – in which one computer acting simply stand-by – to function in the event ev ent of failure of the working computer co mputer
•
Throughout 1970 – Developments in ICs and construction techniques in circuit boards – led to - increment in reliability of the systems – reduction in the cost – increment in the processor power – increase in the fast memory amount – made more correct and dependable softwares to be to write
•
1974 – microprocessors’ advent – made it economical to use distributed computer control system
RTS Definition RTS definition (in as The Oxford dictionary of Computing):
Any system in which the time at which the output is produced is significant. This is usually because the input corresponds to some movement in the physical world, and the output has to relate to that same movement. The lag from input time to output time must be sufficiently small for acceptable timelines
RTS definition (Cooling – 1991):
Real-time systems are those which must produce correct responses within a definite time limit. Should computer responses exceed these time bounds then performance degradation and/or malfunction results
RTS definition (alternate definition):
A real-time real-time system reads inputs from the plant and sends control signals to the plant at times determined by plant operational considerations – not at times limited by the capabilities of the computer system
RTS definition:
A program for which the correctness of operations depends both on the logical results of the computations and the time at which the results are produced Ex.: Aircraft engine control system – sending response to UNIX workstations
Classification of Real-time Systems Computer is connected to the environment env ironment within which it is working by a wide range of interface devices and receives an sends a variety va riety of stimuli – in the real time systems and embedded computers Ex.: Plant input, plant output, and communication tasks shown in Fig.
They have one common feature – that they are connected by physical devices to processes which are external to the computer. These external processes all operate in their own time scales and the computer co mputer is said to operated I real time if actions carried out in the computer relate to the time scales of the external processes
Synchronization between the external processes and the internal actions (tasks) – carried by the computer:
1. Clock based: Synchronization between the above two - if in depends on passages of time, actual time of day (clock based RTS)
2. Event based: Synchronization between the above two – if in depends events Ex.: Closure of a switch
3. Interactive Systems: Relation between the actions in the computer and the system – is much more loosely defined Requirement, typically – in the form that – a set of operations in the computer should be completed within a predetermined time Majority of the communication tasks – are of this category
Control tasks: Even though – not obviously and directly connected to the external
environment – they need to operated in real time – since, time is usually involved in determining the parameters of the algorithms used. It is useful to divide tasks to be carried out by embedded computers into the interactive categories and characteristics of each class are to be recognized
Types of tasks in real time systems: The three types of tasks are
1. Clock-b Clock-base ased d (Cycli (Cyclic, c, Period Periodic) ic) 2. Event Event-b -bas ased ed (Ape (Aperi riod odic ic)) 3. Inte Intera ract ctiv ivee syst system emss
1. Clock-b Clock-based ased Tasks (Cyclic (Cyclic,, Periodi Periodic) c)
Process of the plant - operate in real time Plant time constant: it’s the measure of the time taken by a plant to respond to a change in input or load and is used as a characteristic of the plant May be measured – in hours for some chemical processes or In ms for an aircraft system
Feedback control: It involves the feedback control It requires the sampling rate to be dependent de pendent on the time constant of the process to be controlled
Sampling rate increases – as the time constant value decreases Synchronization in real time – between the two is – required All the required operations (like measurement, control and actuation – within each sampling interval) – can be able to carry out
Completion of the operation – within the specified time: Dependent on 1. Number of operations to be performed 2. Speed of the computer
Real-time clock: It’s the clock – added to the computer - to have the synchronization. It’s
signal – is used to interrupt the operations o perations of the computer – at some predetermined fixed time interval Computers carry out – plant input, plant output and control tasks in response to the clock interrupt. If the clock interrupt is at a faster rate than the sampling rate – count of each e ach interrupt to be to – until it’s the time to run the tasks.
Different sampling rates: used in larger plants – where tasks are subdivided into groups – for controlling different parts of the plant
Clock interrupt – used frequently – to keep a clock and a calendar – and keep the computer – aware of both the time and the date Clock based tasks (cyclic or periodic tasks) – here, task is run once per time period T (cycle time, T) –or run at exactly T unit intervals
2.
Event-based Tasks (Aperiodic)
Systems in which actions are performed in response to some events and not performed at response at particular times or time intervals Ex.:
1. Turing off a pump - Closing a valve when the level of a liquid tank reaches predetermined value – like the one in - level of fuel in the vehicle fuel tank – reaching the pump nozzle 2. Switching a motor off – in response to the closure of a micro-switch indicating - that some desired position had reached
Used extensively – to indicate – alarm conditions and initiate alarm actions Ex.: Indication of too high a temperature or too great a pressure Includes the requirement– that the system s ystem must respond – within a given maximum time to a particular event
Uses the interrupts – to inform the computer system – that action is required. Smaller systems – use polling - where the computer periodically asks (polls) various sensors – to see if action is required
Aperiodic tasks: Events usually occur at non-deterministic intervals Aperiodic tasks – may have deadlines – expressed in terms of having start times or finish times or even both. Ex.: Task Task may be required to start – within 0.5s of an event occurring Task may have to produce p roduce an output – within 0.5s of an event
3. Inte Intera ract ctiv ivee syst system emss
Probably represent – the largest class of real time systems Ex.: 1. Automatic bank tellers
2. Reservation systems for hotels, airlines and car rental co mpanies 3. Computerized tills
Requirement of interactive systems: can be expressed in terms such as ‘the average response tome must not exceed…’ Ex.: Automatic bank teller system might – require an average response time – not exceeding 20s
Event-based systems Vs Interactive systems
The two are 1.
Same: since, it apparently response to a signal from the plant (in this case usually
a person) 2.
Different: Since, it responds a a time determined by the the internal state of the the
computer and without any reference to the environment Ex.: Automatic Automatic bank teller - does not know that – you will miss a train, or that it’s it’s raining hard and you are getting wet, but it’s response – depends on how bus y the communication lines and central computers are and also your amount
Clock-based systems Vs Interactive systems
The two are 1.
Same: since, are capable of displaying the date and time, and also they have a real
time clock – which enables them the m to keep track of time 2.
Different: when the test is done whether or not the answer for the question – ‘Can
the system be tightly synchronized to an external process?’ If answer is ‘yes’ – they are clock-based clock-ba sed If answer is ‘no’ – they are event-based e vent-based
