Chapter 3 E-Commerce Architecture

Chapter 3 E-commerce Architecture

Chapter - 3 E-commerce Architecture

26


3.1. E-commerce Architecture:

Every day more and more business transactions are conducted in the Internet under the umbr umbrel ella la of ECom EComme merc rce. e. The The main main feat featur uree that that dist distin ingu guis ishe hess E-Co E-Comm mmer erce ce from from traditional commerce is the ability to conduct business with spatially distributed partners and products, and achieves considerable reduction in human effort and latency time. The size and complexity of E-Commerce systems make the architecture level of design and specification of the overall system a significant issue. From a good design it is possible to assess the quality and performance of the system before the system is implemented implemented and deplo deploye yed. d. Some Some of the the impo import rtan antt prop proper erti ties es that that can can be asse assess ssed ed from from a desi design gn specificati specification on include include correctnes correctness, s, service service availabili availability, ty, and security security of transactions. This paper discusses the model of E-Commerce architecture towards this goal. There are four types of E-commerce Architecture 1. Clie Client nt Server Server Arc Archi hite tect ctur uree 2. TwoTwo-Ti Tier er Arch Archit itec ectu ture ress 3. Thre Three-T e-Tie ierr Archit Architect ectur ures es 4. Distr Distribu ibuted ted Enterp Enterpri rise se Archit Architect ecture ure 3.2. Client–server architecture:

The client–server architecture is a computing model that acts as a distributed application which partitions tasks or workloads between the providers of a resource or service, called servers, and service requesters, called clients.

[Fig: Client Server architecture]

27


Often clients and servers communicate over a computer network on separate hardware, but both client and server may reside in the same system. A server machine is a host that is running one or more server programs which share their resources with clients. A client does not share any of its resources, but requests a server's content or service function. Clients therefore initiate communication sessions with servers which await incoming requests. A network architecture in which each computer or process on the network is either a client or a server. Client Server architecture use 3 components1. Clients (Clients are Applications): Applications that run on computers and Rely on servers for Files, Devices, Processing power. Make requests, Format data on the desktop. Example: E-mail client an application that enables you to send and receive e-mail. 2. Servers (Servers Manage Resources): Computers or processes that manage network resources, Disk drives (file servers), Printers (print servers), Network traffic (network servers).Store and protect data, Process requests from clients. Example: Database Server -A computer system that processes database q ueries. 3. Communication Networks: Networks Connect Clients and Servers.

3.2.1. Client/Server architecture Propertis:

E-commerce is based on client/ server architecture 1. Client processes requesting service from server processes 2 . F ir st u se d i n 1 98 0s , t he m od el i mp ro ve s t o b e e -c om me rc e usability, flexibility, interoperability and scalability. •

I n e -c o m me r c e th e c li e n t i s d ef i ne d a s th e r e qu e s to r o f a s e rv i c e an d a server is the provider of the service 3 . B r o ws e r i s t h e c li e n t a n d t he c u s t om e r, t h e c o m p ut e r t h at s e n d s the HTML files is the server 4. The server can also be a computer program that provides services to other computer programs

•

A w e b s er v er i s t h e c om pu te r p r og ra m t ha t s er ve s r eq u es te d H T ML pages or files. 5. Uses client/server model and http(hypertext transfer protocol)

28


6 . E v er y c o m pu te r o n t he i nt er ne t t h a t c o nt ai ns a w e b s i te m us t have a web server program. •

M os t p o p ul ar w eb s er ve rs a re D e er fi el d’ s W e b Si te a n d M i cr os of t’ s Internet Information Server (IIS)

•

W e b se r v er s ar e i nc l ud e d a s p a r t o f a la r g er p a ck a g e of i n te r n et a n d intranet related programs for serving e-mail, downloading requests for FTP files and building and publishing web pages.

•

T y p ic a l ly t h e e- c o mm e rc e c u st o me r i s t h e cl i en t a n d t h e bu s in e s s i s the server.

•

I n t he c l ie n t / s e rv e r mo d e l s i ng l e ma c h in e c a n be b o th c l ie n t a nd t h e server

•

T h e c li e n t/ s e rv e r mo d e l ut i li s es a d a ta b a se s e rv e r in w h i ch R D BM S user queries can be answered directly by the server

•

T h e c li e nt / s e rv e r a rc h it e c tu r e r e du c e s n et w o rk t r a ff i c b y pr o v id i n g a query response to the user rather than transferring total files.

•

T he c li en t/ s er ve r m o de l i m pr ov e s m ul ti - us er u p d a ti ng t hr ou g h a graphical user interface (GUI) front end to the shared database.

•

In

client/

server

architectures

client

and

server

typically

communicate through statements made in structured query language (SQL).

3.2.2. Clients-server interaction:

The client–server characteristic describes the relationship of cooperating programs in an application. The server component provides a function or service to one or many clients, which initiate requests for such services. F u n ct i on s s u ch a s e m a il e x c ha n g e, w e b a c c es s a n d d a ta b a se a c c es s , a r e built on the client–server model. Users accessing banking services from their computer use a web browser client to send a request to a web server a t a b an k. T ha t p ro gr am m ay i n t ur n f or wa rd t he r eq ue st t o i ts o wn d a ta b as e c li en t p ro g ra m t ha t s en d s a r eq u es t t o a d a ta b as e s e rv e r a t another bank computer to retrieve the account information. The balance is r e tu r n ed t o t h e b a n k d a t ab a s e c l i e nt , w h ic h i n t u r n s e r v e s i t b a c k t o t h e web browser client displaying the results to the user. The client–server model has become one of the central ideas of network computing. Many

29


business applications being written today use the client–server model. So do the Internet's main application protocols, such as HTTP, SMTP, Telnet, a n d DNS. T h e i n te r a ct i on b e tw e e n c l ie n t a n d s e rv e r i s o f te n d e s cr i be d using sequence diagrams. The Unified Modeling Language has suppo rt for sequence diagrams. Sp eci fic typ es of cli ent s inc lud e web browsers, email clients, and online chat clients. Specific types of servers include web servers, FTP servers, application servers, database servers, name servers, mail servers, file servers, print servers, and terminal servers. M os t web services are al so ty pe s of se rv er s. 3.2.3. Client/Server with File Servers

In the client/server architecture, a file server is a computer responsible for the central storage and management of data files so that other computers on the same network can access the files. A file server allows users to share information over a network without having to physically transfer files by floppy diskette or some other external storage device. Any computer can be configured to be a host and act as a file server.

[Fig: Centralized Client/Server with File Servers]

30


A file server may be an ordinary PC that handles requests for files and sends them over the network. In a more sophisticated network, a file server might be a dedicated network attached storage (NAS) device that also serves as a remote hard disk drive for other computers, allowing anyone on the network to store files on it as if to their own hard drive.

[Fig: Distributed Client/Server with File Servers]

A p ro g ra m o r m ec h an is m t ha t e n ab le s t he r eq u ir e d p r oc e ss es f or f il e sharing can also be called a file server. On the Internet, such programs often use the File Transfer Protocol (FTP).

3.2.4. Client/Server with Database Servers:

A database server is a computer program that provides database services to other computer programs or computers, as defined by the client–server model. The term may also refer to a computer dedicated to running such a program. Database management systems frequently provide database server functionality, and some DB MS s (e.g., MySQL) relies exclusively on the client–server model for database access. Such a server is accessed either through a "front end" running on the user’s computer which displays requested data or the "back end" which runs on the server and handles tasks such as data analysis and storage. In a master-slave model, database master servers are central and primary locations of data while database slave servers are synchronized backups of the

31


master acting as proxies. Events (violation of integrity constraints, temporal conditions, errors) trigger event handlers-> implicit invocation, blackboards, and event.

[Fig: Client/Server with Database Servers]

A DBMS also offers features for recovery and concur rency control Some examples of Database servers are Oracle, DB2, Informix, Ingres, SQL Server. E v er y s er v er u s es i ts o w n q u er y l og ic a n d s tr uc tu re . T h e S Q L q u er y language is more or less the same in all the database servers. At present the majority of existing client/server-based software is to be found in the area of databases, and it is here that the greatest challenge to any corporation currently lies. Richard Finkelstein, President, Performance Computing

3.2.5. Client/Server with Transaction Processing:

32


T ra n sa c ti on p ro c es si ng i s d e si gn ed t o m ai nt ai n a c o mp u te r s ys te m (typically a database or some modern file systems) in a known, consistent state, by ensuring that any operations carried out on the system that are i n te rd e pe n de n t a re e it he r a ll c om p le te d s uc c es sf ul ly o r a ll c an c el ed successfully. 1 ) A to mi ci ty a) A transaction is an indivisible unit of work. b) An all-or-nothing proposition . c) All updates to a database, displays on the clients’ screens, message queues. d) e.g., salary increase for all 1 million employees or none 2 ) C on si st en cy a ) a t r a ns a c ti o n i s a n i nd i v is i bl e u n i t o f wo r k b) integrity constraints (e.g., mgr.salaray > salary c ) I so la ti on

[Fig: Client/Server with Transaction Processing] d) A transaction’s behavior not affected by other transactions running concurrently.

33


e ) E . g. , re s er ve a s ea t. 3 ) S e r ia l iz a ti o n t e c hn i q ue s a ) D ur ab il it y b) Persistence c) A transaction’s effects are permanent after it commits

3.2.6. Advantages of Client Server Architecture:

1) Centralization: In Client Server Networks there is no central administration, here in this architecture there is a centralized control. Servers help in administering the whole set-up. Access rights and resource allocation is done by Servers. Centralized Server Advantage Having a single server hosting the database of user names and passwords and, at the same time, managing what levels of access individual users and computers can have to specific network resources is a distinct advantage of client server architecture. Fewer support staff are needed to manage centralized security accounts than would be needed if security and resource access had to be configured on each individual computer on the network. This advantage is even greater in networks where users typically access resources from a number of computers, as resource access and log-on permissions would have to be configured for each user, on each computer that they use. 2) Proper Management: All the files are stored at the same place. In this way, management of files becomes easy. Also it becomes easier to find files. 3) Back-up and Recovery possible: As all the data is stored on server its easy to make a back-up of it. Also, in case of some break-down if data is lost, it can be recovered easily and efficiently. While in peer computing we have to take backup at every workstation. 4) Up gradation and Scalability in Client-server set-up: A client server system allows the network to easily adapt to upgrades and new technologies. A new operating system or a software upgrade can be implemented from the server across the network without much downtime or inconvenience to users. Computers and servers to a client server network without major interruptions to the network Changes can be made easily by just upgrading the server. Also new resources and systems can be added by making necessary changes in server. 5) Accessibility: From various platforms in the network, server can be accessed remotely. Accessible client server system can be set up with a remote access. It means users can access the network securely from the outside. Servers that allow remote access offer different type of secure connection like wireless, virtual private network (VPN), or dial-up remote access. While maintaining the security

34


of the network, a client server system provides the users the comfort and convenience of accessing the network from the outside, whether a quick access to check emails or a work from home privilege.

6) As new information is uploaded in database, each workstation need not have its own storage capacities increased (as may be the case in peer-to-peer systems). All the changes are made only in central computer on which server database exists.

7) Security: Rules defining security and access rights can be defined at the time of set-up of server. Client-server arrangements aid security efforts because there are fewer servers, limiting the number of people with access to them and increasing security features of the servers.

8) Servers can play different roles for different clients.

3.2.7. Disadvantages of Client Server Architecture:

1. Congestion in Network: Too many requests from the clients may lead to congestion, which rarely takes place in P2P network. Overload can lead to breaking-down of servers. Wherever a single server is used to manage resources on a network, there is always the chance that a failure of the server will bring network business to a halt. For example, if a single server hosts the user names and passwords database for a network, and the server suffers a failure, no users will be able to log on to the network until the server is fixed or replaced. A network where log-on information is configured and held on each computer does not have this problem, as the failure of a single computer does not prevent users of other computers from logging on and accessing network resources. Only the failed computer is affected.

2. Client-Server architecture is not as robust as a P2P and if the server fails, the whole network goes down. Also, if you are downloading a file from server and it gets abandoned due to some error, download stops altogether. However, if there would have been peers, they would have provided the broken parts of file.

3. Dependence: The client-server network model relies on a functioning and available centralized server. If the centralized server is removed from the system or goes down due to problems, the entire network cannot function. However, many client server networks now have backup servers to provide support when a server is lost.

35


4. Cost: It is very expensive to install and manage this type of computing. Typically, the central server computer must be powerful enough to maintain and share resources with the other computers on the network. This entails a substantial cost. A client/server network can be expensive to implement and maintain. First, at least one server is required to create a client/server network. This requires server hardware and software, a server operating system and appropriate licenses to allow the end users to use the network software. If data will be stored centrally, a backup system is needed, which requires backup server software and backup media. Maintaining a client/server network also requires at least one network administrator, which translates into an additional salary. For a smaller network, cost alone may be the deciding factor in opting for a different type of network architecture.

5. Maintenance Client-server networks often requires a staff with at least a single network administrator to manage and maintain the equipment and the network. Other network operating systems, such as peer-to-peer network systems, do not require a network administrator to maintain machines, as this work is distributed among individual clients and their related machines.

6. You need professional IT people to maintain the servers and other technical details of network. Personnel a client/server network requires experienced network personnel to maintain the server, manage security and backup systems, and recover quickly from unexpected outages. Depending on the size and complexity of the client/server network, this could require network administrators, IT security professionals, and/or other IT professionals. A smaller network may be able to combine several of these roles into one position, but additional training is often required as a result.

7. Complexity: A client/server network is naturally more complex, which can be both an advantage and a disadvantage. For instance, data stored remotely can be more secure than data stored locally, but more configurations is required both to ensure the security of that data and to allow the appropriate users to have appropriate access to that data. The very nature of a client/server environment creates challenges in the delivery and accessibility of remote resource.

8. Downtime: When a computer in a simple peer-to-peer network goes down, the only users affected are the user of that particular computer, and perhaps any users who are trying to access data stored on that computer. In contrast, when a server in a client/server network goes down, all users are affected. If security is maintained centrally on that server, workstations are unable to authenticate login

36


information, and users are unable to gain access to their own systems. If user data is stored centrally, users are unable to access their data. If applications are managed centrally, users are unable to run their applications. Any of these scenarios can result in significant loss of productivity and/or revenue.

3.3. Two-Tier Architectures:

A two-tier application generally includes a Java client that connects directly to the database through TopLink. The two-tier architecture is most common in complex user interfaces with limited deployment. The database session provides TopLink support for two-tier applications. Although the two-tier architecture is the simplest TopLink application pattern, it is also the most restrictive, because each client application requires its own session. As a result, two-tier applications do not scale as easily as other architectures. Two-tier applications are often implemented as user interfaces that directly access the database They can also be non-interface processing engines. In either case, the two-tier model is not as common as the three-tier model.

Client User Interface

Server (Business Rules)

(Business Rules)

Data Access

[Fig: Two Tier Client Server] Two-tier architectures have 2 essential components •

A Client PC and

•

A Database Server

3.3.1. Two-tier Considerations: •

Client program accesses database directly o

Requires a code change to port to a different database

o

Potential bottleneck for data requests

37


o

•

High volume of traffic due to data shipping

Client program executes application logic o

L im it ed

by

p ro ce ss in g

c ap ab il it y

of

c li en t

w or ks ta ti on

(memory, CPU) o

R e qu ir es a p pl ic at io n c o de t o b e d is tr ib u te d t o e a ch c li en t workstation.

•

T he u s e r s ys te m i n te rf a ce i s u s u al ly l o c at ed i n t he u s e r’ s d e sk to p environment and the DBM services are usually in a s erver that is a more powerful machine that services many clients.

[Fig: Two Tier Architecture] •

I t r u ns t h e c l i en t p r o ce s se s s e pa ra te ly f r o m t h e s e rv e r p r oc e ss es , usually on a different computer:

38


1 . T h e c l i en t p r o c es s es p r o v i de a n i n te r fa c e f o r t h e c u s to m e r, a n d gather and present data usually on the customer’s computer. This part of the application is the presentation layer 2 . T h e s e r ve r p r o ce s s es p r o v id e a n i n t e rf a ce w i t h t h e d a t a s t or e o f the business. This part of the application is the data layer 3 . T h e b u si ne ss l og ic t ha t v a li da t es d a ta , m on it or s s e cu ri ty a nd permissions, and performs other business rules can be housed on either the client or the server, or split between the two. 1 . F u n da m e nt a l u n i ts o f w o r k r e q u ir e d t o c o m p le t e t h e b u s in e s s process 2. Business rules can be automated by an application program •

T yp ic al ly us ed in e- co mm er ce 1 . I n te r n et r e t ri e v al , d e s ic i o n s up p o rt

•

U s ed i n di s tr i bu t e d co m p ut i ng w h e n t h er e a re f e we r t ha n 1 0 0 p e o pl e simultaneously interacting on a LAN.

•

I mp le m en t at io n o f p r o ce ss in g m a n ag e me n t s e rv ic e s u s in g v e n do r proprietary db procedures restricts flexibility and choice of RDBMS for applications.

•

A ls o l ac ks f le xi bi li ty i n m ov in g p ro gr am f un ct io na li ty f ro m o ne server to another.

3.3.2. The advantages of the two tier architecture: •

•

Development Issues: Fast application development time, Easy to setup and maintain available tools are robust and lend themselves to fast prototyping to insure user needs a met accurately and completely. Conducive to environments with homogeneous clients, homogeneous applications, and static business rules. Performance: Adequate performance for low to medium volume environments Business logic and database are physically close, which provides higher performance.

3.3.3. The disadvantages of the 2-tier architecture:

39


•

•

•

•

•

•

•

S c al ab il it y: k e y c o nc e rn w it h t he 2 - ti er m od e l i s s ca la b il it y. Application performance can be expected to degrade rapidly when the number of concurrent users reaches a threshold between a few hundred and one thousand users. This is true even for large database s er ve rs . T he c hi ef r ea so n i s t ha t e ac h c li en t r eq ui re s i ts o wn connection and each connection requires CPU and mem ory. As the n um be r o f c on ne ct io ns i nc re as es , t he d at ab as e p er fo rm an ce degrades. Poor Logic Sharing: Traditional two-tier architectures keep business l o gi c o n t h e c l ie n t. W h e n l o g ic i s i n t h e c l i e nt , i t i s u s ua l ly m o re difficult to re-use logic between applications and amongst tools. Application Distribution: Application changes have to be distributed to each client. When there are a large number of us ers, this entails considerable administrative overhead. Remote Usage: Remote users (e.g. customers), probably do not want to install your application on their clients-they would prefer "thin" c li en ts w he re m in im al ( or n o) c li en t s of tw ar e i ns ta ll at io n i s required. D a ta b a se S t ru c tu r e : o t he r a p p li c at i on s t h at a c c es s y o u r d a t ab a s e w il l b e co m e d e pe n de n t o n t h e e x is ti ng d at ab a se s tr uc tu r e. T hi s means that it is more difficult to redesign the database since other applications are intimate with the actual database structure. Development Issues: Complex application rules difficult to implement in database server – requires more code for the client. Complex application rules difficult to implement in client and have poor performance. Changes to business logic not automatically enforced by a server – changes require new client side software to be distributed and installed. Not portable to other database server platforms. Performance: Inadequate performance for medium to high volume environments, since database server is required to perform business logic. This slows down database operations on database server.

40


3.4. Three-Tier Architecture(Multitier architecture):

Multi-tier architecture (often referred to as n-tier architecture) is a client– server architecture in which the prese ntatio n, the applic ation proce ssing , and the data management are logically separate processes. For example, an application that uses middleware to service data requests between a user and a database employs multitier archit ecture . [Fig: Three Tier Architecture]

T h e m os t w id e sp re a d u se o f m ul ti -t ie r a rc h it ec tu r e i s t he t hr e e- ti er architecture. Three-tier is a client–server architecture in wh ic h th e user interface, functional process logic ("business rules"), computer data storage and data access are developed and maintained as independent modules, most often on separate platforms. It was developed by John J. Donovan in Open Environment

41


Corporation (OEC), a tools company he founded in Massachusetts. The three-tier model is software architecture and a pattern. Apart from the usual advantages of modular software with well-defined interfaces, the three-tier architecture is intended to allow any of the three tiers to be u pg rade d o r re place d ind epe nd ently in respo nse to ch ang es in requirements or technology. For example, a change of operating system in the presentation tier would only affect the user interface code. Typically, the user interface runs on a desktop PC or workstation and uses a standard graphical user interface, functional process logic may consist of one or more separate modules running on a workstation or application server, and an RDBMS on a database server or mainframe conta ins the compu ter data storage logic. The middle tier may be multi-tiered itself (in which case the overall architecture is called an "n-tier architecture").

3.4.1. Three-tier of the architecture:

N-tier application architecture provides a model for developers to create a flexible and reusable application. By breaking up an application into tiers, developers only have to modify or add a specific layer, rather than have to rewrite the entire application over. There should be a presentation tier, a business or data access tier, and a data tier. The concepts of layer and tier are often used inter changeably. However, one fairly common point of view is that there is indeed a difference, and that a layer is a logical structuring mechanism for the elements that make up the software solution, while a tier is a physical structuring mechanism for the system infrastructure. T hr ee -t ie r ( la ye r) i s a c li en t- se rv er a rc hi te ct ur e i n w hi ch t he u se r i n te r fa c e , b u s in e s s p r oc e ss ( b us i ne s s r u le s ) a n d d a t a s t or a g e a n d d a ta a c c es s a r e d e v e l o pe d a n d m a i n t ai n ed a s i n de p e nd e n t m o d u l es o r m o s t often on separate platforms. Basically, there are 3 layers: Tier 1 (presentation tier, GUI tier), tier 2 (business objects, business logic tier) and tier 3 (data access tier). •

P re se nt at io n t ie r O r P re se nt at io n L ay er ( UI ): Th is i s t he t op mo st level of the application. The presentation tier displays information related to such services as browsing merchandise, purchasing, and

42


s ho pp in g c ar t c on te nt s. I t c om mu ni ca te s w it h o th er t ie rs b y outputting results to the browser/client tier and all other tiers in the network. Presentation layer contains pages like .aspx or windows form where data is presented to the user or input i s taken from the user. A local computer on which either a Web browser displays a Web page that can display and manipulate data from a remote data source, or (in non-Web-based applications) a stand-alone compiled front-end application •

Application tier ( business logic, logic tier, data access tier, or middle tier): The logic tier is pulled out from the presentation tier and, as it s ow n l ay er ; i t c on tr ols a n a pp lic at io n’ s f un ct io na li ty b y performing detailed processing. BAL contains business logic, validations or calculations related with the data, if needed. I will call it Business Access Layer in my demo. A Server computer that h o s ts c o mp o n en t s w h ic h e n c ap s u la t e a n o r g an i z at i on ' s b u s in e ss r u le s . M i dd l e- t ie r c o m po n e nt s c a n e i th e r b e A c ti v e S e r ve r P a g e scripts executed on Internet Information Server, or (in non-Web based applications) compiled executables.

•

Data A cce ss Laye r (DAL) O r Data tie r: This tier co nsists o f database servers. Here information is stored and retrieved. This tier k e ep s d at a n e ut ra l a n d i nd e pe nd e nt f ro m a pp li ca t io n s er ve rs o r business logic. Giving data its own tier also improves scalability and performance.DAL contains methods that helps business layer to connect the data and perform required action, might be returning data or manipulating data (insert, update, delete etc).A computer h os ti ng a d ata ba se m an ag em en t s ys te m ( DB MS ), s uc h a s a M ic r os of t S QL S e rv e r d a ta b as e . ( In a t wo -t ie r a p pl ic a ti on , t he middle-tier and data source tier are combined.)

3.4.2. Component & Consideration of 3tier architecture:

Three Tier architectures have 3 essential components: •

A Client PC

•

A n A pp lic at io n S er ve r

•

A Database Server

Three-Tier Architecture Considerations:

43


• Client program contains presentation logic only o

o

o

Less resources needed for client workstation No client modification if database location changes Less code to distribute to client workstations

• One server handles many client requests o

More resources available for server program

o

Reduces data traffic on the network

[Fig: 3 Tier Architecture]

Also called as multi-tier architecture.A middle tier is added between the client environment and the DBM server environment.

3.4.3. Implemantation of 3 Tier Architecture:

Variety of ways to implement three tier architectures : 1 . T r a ns a c ti o n p r oc e s si n g ( TP ) m o ni t or s

44


2 . M es sa ge s er ve rs 3 . A p pl ic a ti on s e rv er s

3.4.3.1.Three-Tier Architectures with TP Monitor: •

The most basic type o

•

•

•

•

•

Type of message prioritisation service

queuing,

transaction

scheduling,

Client connects to TP instead of the DB server T he t ra n sa ct io n i s a c ce p te d b y t he m on it or w h ic h q u eu e s i t a n d takes responsibility to complete it by freeing up the client When a third part provides this service it is called TP heavy When it is embeded in the DBMS, it can be considered 2-tier and is referred to as TP lite TP monitor provides: o

o

The ability to update multiple DBMSs in a single transaction Connctivity to a variety of data sources (flat files & nonRDBMSs)

o

The ability to attach priorities to transactions

o

Robust security

•

More scalable than a 2-tier architecture

•

Most suitable for e-commerce with many thousands of users

3.4.3.2. Three-Tier Architectures with Message Server:

•

M e s sa g e s a r e p r io r it i se d an d pr o c es s e d a s y nc h ro n o us l y

45


1. Headers contain priority info, the address, the id no •

M e ss ag e s er ve r c on n ec ts t o t he R D BM S a nd o t he r d at a sources

•

T he m es sa ge s er ve r f o cu se s o n i nt el li ge nt m es sa ge s, whereas the TP environment has the intelligence in the monitor and treats transactions as dumb data packets

•

T he y a re s ou nd b us in es s s ol ut io ns f or t he w ir el es s infrastructures of m-commerce.

3.4.3.3. Three-Tier Architectures with an Application Server:

1 . A ll oc at es t he m a in b od y o f a n a p p li ca ti on t o r u n o n a s h a re d h o s t rather than in the user system interface client environment 2. The application server does not drive GUIs rather it shares business logic, computations, and a data retrieval engine. 3 . W it h le ss s w on t he c l ie n t 1 . T h e re i s l es s c on c e rn w i th s e c ur i ty , 2 . A p p li c a ti o n s a r e mo r e sc a la b le 3 . S u pp o rt a n d i n st al la ti on c o st s a r e l es s o n a s in g le s er ve r t h a n maintaining each on a desktop client.

3 .4 .3 .4 . T hr ee -T ie r A rc hi te ct ur es w it h a n O bj ec t R eq ue st B ro ke r Standard:

46


•

•

Need for improving interoperability and object request broker (ORB) standards in the client/ server model. O RB s up po rt i n a n et wo rk o f c li en ts a nd s er ve rs o n d if fe re nt computers means o

o

•

•

•

A client program (object) can request services from a server program Object without having to understand where the server is in a d is tr ib ut ed n et wo rk o r w ha t t he i nt er fa ce t o t he s er ve r program looks like

ORB is the programming that acts as the mediary or as a broker between a client request for a service from a distributed object or component and server completion of that request. There are two prominent distributed object technologies: o

Common Object Request Broker Architecture (CORBA)

o

Component Object Model (COM)

The industry is working on operability between CORBA and COM

3.4.4. Advantages of 3 tier architecture:

a . T h e a d va n ta g e o f a 3 - t ie r s y s te m i s t h at t h e c o n te n ts o f a n y o f t h e tiers/layers can be replaced without making any resultant changes in any of the others. For example: A change from one D BMS to other will only involve a change to the part in the data access layer. A change in the Use Interface (from desktop to the web, will need only some changes in the components of the presentation layer. b. The benefit of writing the presentation and business layered architecture in different languages is that it is a n advantage of the presentation and business layers is that it is feasible to use different developer teams to work on each. It means that only PHP skills are required for data access layer and business layer, and HTML, CSS a nd X LS s ki ll s f or p re se nt at io n l ay er s. I t i s e as ie r t o f in d a

47


d ev el op er w it h s ki ll s i n o ne o f t he se r at he r h av in g a k in d o f developer having all of the skills. c . A n ot he r m ai n a dv a nt ag e o f u si ng X M L/ XL S i n t h e p re se n ta ti on layer is that it the output can be changed from HTML to WML or PDF or any other format using a different XLS style sheet. XLS f il es c an b e u se d t o c ha ng e X ML d oc um en ts i nt o a n um be r o f formats and not only HTML. d . I t h a s b e t te r w a i t b a l an c in g s y s t em b e c a u se t h e e n t i re w o r k l o a d i s divided. e. Security polices can be imposed without effecting the clients. f . S c a la b il i ty : T h e k e y 3 - t i er b e n e f it i s i m p r ov e d s c a la b i li t y s i nc e t h e application servers can be deployed on many machines. Also, the database no longer requires a connection from every client -- it only requires connections from a smaller number of application servers. In addition, TP monitors or ORBs can be used to balance loads and dynamically manage the number of application server(s) available. g.

Better Re-use: The same logic can be initiated from many clients or applications. If an object standard like COM/DCOM or CORBA is e mp lo y ed ( a s d is cu ss e d i n t oo l d e pe nd e nc e) , t h en t he s pe c if ic language implementation of the middle tier can be made transparent.

h . I mp r ov e d D a t a I n t eg ri ty : s in c e a l l u p da te s g o t hr o ug h t he m id d le tier, the middle tier can ensure that only valid data is allowed to be updated in the database and the risk of a rogue client application corrupting data is removed. i. Improved Security: Security is improved since it can be implemented at multiple levels (not just the database). Security can be granted on a service-by-service basis. Since the client does not have direct access to the database, it is more difficult for a client to obtain unauthorized data. Business logic is generally more secure since it is placed on a more secure central server. j. Reduced Distribution: Changes to business logic only need to be updated on the application servers and do not have to be distributed to all the clients.

48


k. Improved Availability: mission-critical applications can make use of redundant application servers and redundant database servers. With redundant servers, it is possible to architect an application so that it can recover from network or server failures. l.

Hidden Database Structure: since the actual structure of the database is hidden from the caller, it is possible that many database changes can be made transparently. Therefore, a service in the middle tier that exchanges information/data with other applications could retain its original interface while the underlying database structure was enhanced during a new application release.

m. Development Issues: Complex application rules easy to implement in application server. Business logic off-loaded from database server and client, which improves performance. Changes to business logic a ut om at ic al ly e nf or ce d b y s er ve r – c ha ng es r eq ui re o nl y n ew application server software to be installed. Application server logic i s p or ta b le t o o th e r d a ta ba s e s er ve r p la tf or ms b y v ir tu e o f t he application software. n . P e rf or ma n ce : S u pe ri or p e rf or ma n ce f or m ed iu m t o h ig h v o lu m e environments

3.4.5. The disadvantage of the 3-tier architecture:

a. Increased Complexity/Effort: In general, it is more difficult to build a 3-tier application compared to a 2-tier application. The points of communication are doubled (client to middle tier to server, instead of simply client to server) and many handy productivity enhancements provided by client tools (e.g. Visual Basic, PowerBuilder, Delphi) will be foregone or their benefit will be reduced with a 3-tier architecture. b. Fewer Tools: There are many more tools available for a 2-tier model (e.g. most reporting tools). It is likely that additional programming effort will be required to manage tasks that an automated tool might handle in a 2-tier environment. Creates an increased need for network traffic management, server load balancing, and fault tolerance. Current tools are relatively immature and are more complex. Maintenance tools are currently inadequate for maintaining server libraries. This is a potential obstacle for simplifying maintenance and promoting code reuse

49


throughout the organization. There is more processing on the web server, It does not interact with the WMS server directly. c. More complex structure more difficult to setup and maintain the physical separation of application servers containing business logic functions and database servers containing databases may moderately affect performance. d. Development Issues: For More complex structure it is mo re di ff ic ul t to se tu p and maintain. e. Performance: The physical separation of application servers containing business logic functions and database servers containing databases may moderately affect performance.

3.5. Distributed Enterprise Architecture:

50


T he D is tr ib u te d E nt er p ri se A rc h it ec tu r e h a s b ee n d e si gn e d t o support distribution of key components. For instance, the Data Adapters c a n b e a n d i n f ac t w e re d is tr ib u te d t o o th e r m ac h in e s t o s up p or t t he Distributed Enterprise Architecture effort on IL’03. The notion here is that it is easy to use different devices on the network to do data collection so that a) we can avoid overloading any given device and b) we can place bandwidth intensive Data Adapters on the same LAN (and even the same virtual LAN) as other important devices to minimize ill effects of highly distributed systems. Because the system is web-based, anyone anywhere o n t he n et wo rk (c on ne cte d v ia W AN o r L AN a nd p os se ss in g t he a pp ropriate p rivilege s) c an lo g on to the Distribu te d En te rp rise A r c hi t ec t ur e P o r ta l ( d e sc r ib e d b e lo w ) , r u n i n te r a ct i v e q u e ri e s a g a in s t D is tr ib u te d E nt er pr is e A rc h it ec tu r e d a ta b as e s a n d v ie w , r e fr e sh , o r publish analysis products.

[Fig: Logical diagram of an Enterprise Architecture]

The design has further been guided by the need for the Data Collection component to be real-time and for the analysis component to at least be

51


n e ar r ea l- ti me . T h e D at a A da p te r f ra me w or k i nc o rp o ra te s r ea l- ti me control and exception management. The multiple layers (e.g. separating t h e d a t a l a ye r f r om t h e i n te g r at i on / se r v ic e s l a y er ) a l lo w t h e a n a ly s is c om po ne nt t o b e a bl e t o y ie ld l im it ed r es ul ts i n r ea l t im e o r m or e complete results at near real time or after action. Currently, a complete Distributed Enterprise Architecture system can be run on a single highend laptop. As the load placed on the Distributed Enterprise Architecture s ys te m i nc re as es , e ith er th ro ug h t he e xp an sio n o f t he c oll ec tio n r eq uir em en ts o r th ro ug h in cr ea se d u se r tr aff ic s up po rte d b y t he Distributed Enterprise Architecture Portal, additional servers are needed to distribute the load so that the Data Adaptors do not fall behind and the P or ta l r em ai ns r es po ns iv e A g ui di ng p ri nc ip le i n t he d es ig n o f t he Distributed Enterprise Architecture system is to separate functionality, Implemented by code, from configuration.

3.5.1. Basic Properties of Distributed Enterprise Architecture:

Distributed Enterprise Architecture collection parameters are typically s p e ci f ie d i n c o ll e c ti o n s t ra t eg i e s, a n d s y st e m p a ra m e te r s a r e s t or e d i n name-value pairs, all of which is stored in a central database. •

•

•

•

Distributed Enterprise Architecture based on ORB technology. Distributed Enterprise Architecture uses shared, reusable business models on a business enterprise-wide scale. Distributed Enterprise Architecture is standardised business object m od e ls a n d d is tr ib u te d o b je ct c o mp u ti ng a re c o mb in e d t o g iv e greater flexibility to the business. D is tr ib ut ed E nt er pr is e A rc hi te ct ur e w it h t he e me rg en ce a nd popularity of ERP sw, distributed enterprise architecture promises to enable e-commerce to extend business processes at the enterprise level.

52


[Fig: critical components of any distributed enterprise architecture]

3.5.2. The advantages of the Distributed Enterprise Architecture:

1 . A l lo w s s i g n if i ca n t m o d i fi c a ti o n o f D C A d e p lo y m en t w i t h o ut c o d e changes. 2. Non-developers can configure the DCA and the Data Adapters. 3 . C o ll e ct i o n S t r a te g i es a n d s y s t em p a r am e te r s c a n b e c o n f i gu r at i on managed separately from the code base. 3.5.3. The Disadvantages of the Distributed Enterprise Architecture:

1. Software implementation is difficult. 2. Network setup is complicated. 3. More components to fail because of so many components are used. 4. Security issue is low to maintain.

53

Chapter 3 E-Commerce Architecture

Recommend Documents