A server farm is a collection of computer servers usually maintained by an enterprise to accomplish server needs far beyond the capability of one machine. Often, server farms will have both a primary and a backup server allocated to a single task, so that in the event of the failure of the primary server, a backup server will take over the primary server's function.

Server farms are typically co-located with the network switches and/or routers which enable communication between the different parts of the cluster and the users of the cluster.

Server farms are commonly used for cluster computing. Many modern supercomputers consist of giant server farms of high-speed processors conncted by either Gigabit Ethernet or custom interconnects such as Myrinet.

Another common use of server farms is for web hosting.

Server farms are increasingly being used instead of or in addition to mainframe computers by large enterprises, although server farms do not as yet reach the same reliability levels as mainframes. Brecause of the sheer number of computers in large server farms, the failure of individual machines is a commonplace event, and the management of large server farms needs to take this into account, by providing support for redundancy, automatic failover, and rapid reconfiguration of the server cluster.

The performance of the very largest server farms (thousands of processors and up) is typically limited by the performance of the data center's cooling systems rather than by the performance of the processors. For this reason, the critical design parameter for such systems tends to be performance per watt of generated heat, rather than performance per processor.

See also

• blade server