Object Oriented Software Engineering   View all facts   Glossary   Help

subject > pattern > software architecture > distributed architecture > client-server architecture > three-tier model of client-server architecture

three-tier model of client-server architecture
subject			fact

three-tier model of client-server architecture	has definition A model in which a server communicates with both a client (usually through the Internet) and with a database (usually within an intranet, for security reasons)
	is a subtopic of 9.5 - Architectural Patterns
	is used to design three-tier systems
	is a kind of client-server architecture
client-server architecture	has advantages The computational work can be distributed among different machines The clients can access the server's functionality from a distance The client and server can be designed separately, so they can both be simpler than a program that does everything All the data can be kept centrally at the server, thus making it easier to assure its reliability Conversely, distributing data among many different geographically-distributed clients or servers can mean that if a disaster occurs in one place, the loss of data is minimized The server can be accessed simultaneously by many clients Competing clients can be written to communicate with the same server, and vice-versa
	has alternatives Have a single program on one computer that does everything Avoid having any communication and have each computer perform the work separately Have some mechanism other than client-server communication for exchanging information
	is one of the most widely used ways of structuring software systems
distributed architecture	facilitates designing for flexibility because distributed systems can often be easily reconfigured by adding extra servers or clients and clients and servers can be developed by competing organizations, giving the customer a choice
	facilitates divide and conquer since dividing the system into processes is a very strong way to divide the system up
	increases abstraction because separate distributed components are often good abstractions
	increases reuse because it is often possible to find suitable frameworks on which to build good distributed systems
	reduces coupling since there is usually only one communication channel between distributed components although you have to watch out for control coupling
software architecture	constrains the overall efficiency, reusability and maintainability of the system
	is decided early in the design process, although it will continue to mature as iterative development proceeds
	must be understood by software engineers
	see also software architecture^2
	see also software architecture^3
pattern	should be described in an easy-to-understand form so that people can determine when and how to use it
	should contain a solution that has been proven to effectively solve the problem in the indicated context

Next client-server architecture: peer-to-peer    Up: client-server architecture