Overview
FrançaisRead this article from a comprehensive knowledge base, updated and supplemented with articles reviewed by scientific committees.
Read the articleAUTHORS
-
Michel RIVEILL: Professor at the University of Nice - Sophia-Antipolis
-
Roland BALTER: Professor at Joseph-Fournier University, Grenoble, SIRAC Laboratory
-
Fabienne BOYER: Lecturer at Joseph-Fournier University, Grenoble, SIRAC Laboratory
INTRODUCTION
There are several models for organizing a distributed application. These include models based on message or event/reaction exchange, which are well suited to asynchronous communication; the client-server model, which relies on a well-known linguistic abstraction, the synchronous procedure call; models using code mobility, such as mobile agent systems; and distributed object models, which give the application designer the illusion of a shared memory of distributed objects. In this article, we focus mainly on the client-server model, as it is the most widespread in industrial products today. We will see, however, that the boundary between the various models for organizing distributed applications is not watertight, and that distributed applications built according to the client-server model quite often borrow properties and mechanisms from other models.
The article develops the client-server model along two axes: a "development environment" axis, which presents the tools for building applications, in particular the remote procedure call; a "system" axis, which presents the principles for implementing the remote procedure call in a heterogeneous distributed environment. These system services are generally grouped together in a software layer between the application and the operating system, usually referred to by the generic term "middleware".
The basic client-server model involves a client process, which requests the execution of a service, and a server process, which performs this service. Client and server are located on two machines linked by a communication network. This model was introduced to implement the first distributed applications (file transfer, connection to a remote machine, e-mail, etc.), each implemented by a specific application protocol. In a second stage, a common construct, the remote procedure call, was introduced to provide a general tool for programming client-server applications.
In this article, we have deliberately grouped together the two existing synchronous communication models, linking a client process and a server process: the model derived from procedural programming, enabling remote procedure calls (RPC: Remote Procedure Call), and its adaptation to object-oriented languages, enabling remote method calls (RMI: Remote Method Invocation). These two models share common foundations, and the latter is a natural evolution of the former, imposed by the development of object-oriented languages.
Exclusive to subscribers. 97% yet to be discovered!
You do not have access to this resource.
Click here to request your free trial access!
Already subscribed? Log in!
The Ultimate Scientific and Technical Reference
This article is included in
Software technologies and System architectures
This offer includes:
Knowledge Base
Updated and enriched with articles validated by our scientific committees
Services
A set of exclusive tools to complement the resources
Practical Path
Operational and didactic, to guarantee the acquisition of transversal skills
Doc & Quiz
Interactive articles with quizzes, for constructive reading
Synchronous communication between programs using RPC and RMI
References
Exclusive to subscribers. 97% yet to be discovered!
You do not have access to this resource.
Click here to request your free trial access!
Already subscribed? Log in!
The Ultimate Scientific and Technical Reference