Load Balancing of Execution Servers

Performance Manager uses a static approach to balance the load between execution servers within the same location. This approach implies that load balancing takes place only upon user operations, except when an execution server is no longer available and the failover system triggers. Whenever a monitor is scheduled to be moved to another execution server, the server with the lowest number of tasks is selected.

Server selection takes place whenever one of the following operations happens:

Creating a new scheduled monitor
Defining a schedule for a previously not scheduled monitor
Adding a new location to a schedule
Deactivating an execution server - monitors are shifted to remaining execution servers in the location
Activating an execution server - the new server adopts monitors from existing execution servers
Failover of an execution server - operates equally as deactivating a server

Activating or deactivating a project or monitor does not trigger a new server selection.