A thin-client server is a specially-designed system that handles a larger portion of its clients' operations than a standard server typically does. These servers will deal with operations that most users would typically associate with a local computer, such as saving files, processing information or even holding the operating system. A thin-client server system is often used in situations where a worker's computer doesn't need a huge amount of functionality or power, a situation common in most offices.
In this situation, a client is a terminal that is connected to a server and it often used as a workstation by an employee. There are two basic types of clients: thin and fat. A fat-client contains a full computer at the worker’s location. This client has drives, local software, internal processing power and all the other things typically associated with a computer.
Thin-clients don't have some or any of the parts of a fat-client. Instead of a computer at the worker's desk, there may be anything from a scaled-down system, to an external drive bay or just a simple routing box. This difference in local clutter gives the two client types their names.
Since many of the things used in a standard computer are missing on the terminal side, they are present on the thin-client server side. Even though all the parts are there, they are typically much more efficient when part of the greater server system. For instance, rather than have a full hard drive at every worker's desk, a thin-client server will have a single large drive that is partitioned for each user. The users will each have their own personal spaces, but they are all in the same location.
In addition to providing basic hardware, a thin-client server can do much more. Some of these servers do all of the processing for their terminals. When the user issues a command, such as "Open a spreadsheet," the command goes to the server to execute. The server processes the command and sends the result back to the client. This is especially common when the server also contains the programming used by the system.
This relationship often results in a totally virtual system for the user. While it may seem like the worker is on a normal computer, the entire process is happening somewhere else, and the worker is simply observing and inputting commands. This works very well when a client needs simple computer access, such as email, web browsing or creating standard office documents, but not so well for other things. When a computer needs to perform a strenuous task, such as video editing or virtual design, a standard fat-client usually works better.