Configuring the WSM using Device Manager # Go back one pageGo to the next page#Go to this book's Index

About server load balancing

Server load balancing (SLB) improves network performance by eliminating the potential for overloading servers that provide specialized services. An 8600 with a WSM is aware of the services provided by each server, and uses availability and performance factor metrics to direct and balance session traffic.

In average multiple-server networks without server load balancing, each server usually provides one or two unique services. Servers providing services in high demand are overloaded. This impacts the performance of the entire network as user requests rejected by the server are then resubmitted by user stations. Overutilization of key servers often occurs in networks where other servers are actually available.

SLB maps multiple physical (real) servers providing unique services to one logical (virtual) server connection and applies an algorithm to balance TCP/UDP requests across the server pool.

The figure below illustrates the difference between networks using specialized servers and server load-balanced networks.

By front-ending the server pools, the WSM improves network performance, reliability, and scalability as it intercepts and distributes user requests. When a request is received, the WSM uses load-balancing algorithms to bind the session to the real IP address of the best available resource; and maps the session to the virtual connection to protect the server pool.

Basic SLB for specific services is possible when each server in the pool has access to identical content and is front-ended by the 8600 with WSM. Advanced (or Extended) Layer 7 SLB does not require content to be duplicated. For more information, see

See also:


Go back one pageGo to the next page##Go to this book's Index