The use of ports and their identifying numbers are an extension to the addressing scheme. Once the address is used to deliver data to the wanted host on the network, the port number is used to identify the process for which the data is used. This enables one host to provide more than one service.

How you define the port number depends on your configuration. Some applications make use of standard, or well-known, port numbers. Two applications at the same address cannot use the same port number. If you are configuring your system with multiple instances of TCP/IP on the same system, however, they will have different addresses and therefore the same port number can be used for the same function on each stack.

TCP⁄IP assumes the well-known port number unless you explicitly specify otherwise when entering a TCP⁄IP command. A port number is entered as a decimal number on TCP⁄IP commands. For those cases when you are requesting the services of a user-developed server, you need to know the port number of that server.