It is unusual not to rely on any other computers in order to perform a task. What is more usual is that different tasks are assigned to different servers. The print server will be one machine, a file server will be another and a web server maybe a third. This is done as a server can not service too many requests. If it does then it may start to slow down. Each request takes time and although servers tend to be very powerful computers, they do not have infinite resources. Eventually they will run out.

Servers may need to talk to other servers in order to complete its task. Consider the situation where a web server requires access to a database stored on a separate machine.

So the web server,in order to fulfil a request from the client, must access yet another server. This is known as a three tier architecture. The first tier is the client, the second is the web server and the third is the database server. It is possible to have four tier or even more, depending on the task which needs doing.