In my last question, I explained that Worker Groups are collections of XenApp servers that are managed as a single unit. Worker Groups often contain servers of the same configuration. Using similar configurations, an administrator can scale out application availability by adding more XenApp servers to a Worker Group as user load increases.

Servers in those Worker Groups are often also located in the same physical location, although they don't necessarily need to be. It's a good idea for users to connect to the XenApp server that's geographically appropriate for their location and their needed applications. A Worker Group preference and failover policy provides a way to link users to that appropriate Worker Group.

Finding the appropriate Worker Group during a service outage is also important. Sometimes a Worker Group or the network connection between a client and its primary Worker Group isn't available. When this happens, users must be re-routed to the next most appropriate Worker Group to obtain their applications. A Worker Group Preference and Failover policy identifies the primary and backup Worker Groups to which users will connect in any of these situations.