Client-server application model

Typically, the client and server components are on different computer systems, and the communication mechanism is based on some form of networking technology, but it is possible to have both the client and server on the same host simply communicating between processes.

To make the development of client-server applications easier, an application programming interface (API) is used. Typically, the server is designed first, and its functionality is structured into a set of commands. These commands are programmed into an API, which becomes the language for interacting with the server. When a client wants the server to do something, it makes a request through the API. The API handles all of the communications and operating system functions, gets the server to perform the task, and returns the appropriate information to the client. When the client software is developed, the programmer need not know any of the details of the server environment or communications mechanisms.

The AR System server has a fully defined API that is common to all server platforms, both Windows and the UNIX platforms. The AR System clients use this API to interact with the servers. A client on any platform can work with a server on any platform. As long as a client can connect to an AR System server, it can communicate all its requests and receive the replies using this common language. The client need not know the platform the server is running or other clients that are using the same server.

The AR System API is defined by a strict set of programming functions and data structures, which are documented in AR-System-C-API-functions and Data-structures. The API is implemented as a C library and associated files that can be linked into your programs. Third-party programs linked to the AR System API library become clients to an AR System server.