All Unix/Linux systems provide TCP/IP support for network programming. In this section, we shall explain the platforms and the server-client computing model for network programming.
1. Network Programming Platforms
In order to do network programming, the reader must have access to a platform that supports network programming. Such a platform is available in several ways.
(1). User Account on Server Machines: Nowadays, almost all educational institutions provide network access, often in the form of wireless connections, to their faculty, staff and students. Every member of the institution should be able to login to a server machine to access the Internet. Whether the server machine allows general network programming depends on the policy of the local network administration. Here, we describe the setup of a network programming platform at the author’s institution, EECS of Washington State University. The author maintains a private server machine
cs360.eecs.wsu.edu
The server machine runs Slackware Linux version 14.2, which comes with a full compliment of support for network programming. The server machine is registered with the DNS server of EECS of WSU. When the server machine boots up, it uses DHCP (Dynamic Host Configuration Protocol) to obtain a private IP address from the DHCP server (RFC 2131 1997). Although not a public IP address, it can be accessed on the Internet through NAT (Network Address Translation). Then the author creates user accounts for students in the CS360 class for them to login. Students typically connect their laptops to the Internet via the WSU wireless network. Once on the Internet, they can login to the cs360 server machine.
(2). Standalone PCs or Laptops: Even if the reader does not have access to a server machine, it is still possible to do network programming on a standalone computer by using the localhost of the computer. In this case, the reader may have to download and install some of the network components. For example, Ubuntu Linux users may have to install and configure the Apache server for HTTP and CGI programming, which will be described later in Sect. 13.17.
2. Server-Client Computing Model
Most network programming tasks are based on the Server-Client computing model. In the Server- Client computing model, we run a server process at a server host first. Then we run a client from a client host. In UDP, the server waits for datagram from a client, processes the datagram and generates a response to the client. In TCP, the server waits for a client to connect. The client first connects to the server to establish a virtual circuit between the client and the server. After the connection is made, the server and client can exchange continuous streams of data. In the following, we shall show how to do network programming using both UDP and TCP.
Source: Wang K.C. (2018), Systems Programming in Unix/Linux, Springer; 1st ed. 2018 edition.