TCP/IP负载均衡

TCP/IP load balancing is a mechanism for distributing many incoming |service requests between two or more service providers. The distribution is|adjusted by a load balancing algorithm. The effect is to improve service response |times and reliability. TCP/IP load balancing is also called Network Dispatch, or IP |spraying.

|

Two or more servers are collected together in a single configuration. This |configuration is often called a group or a cluster, and is identified by a |single host name or TCP/IP network address. Any application or client requiring |a service sends the request to the single host name. The request is passed |to one of the servers within the group, which performs the task and responds |as normal. A load balancing algorithm selects which server will respond from|within the group. Various algorithms may be used. The performance of a Gateway group |is affected by the number of servers within the group, and the algorithm chosen |for choosing the server to receive a request.

|

For more information on TCP/IP load balancing on z/OS, see the Redbook Communications Server for z/OS V1R2 TCP/IP Implementation Guide, SG24-6517-00.

微软上说明：http://support.microsoft.com/kb/303817