IWMSLIM – Application Environment Limit Service

The IWMSLIM service should be used to tell WLM the total number of server instances which are supported by the application. WLM will ensure that no more server instances will be started in the system.

In addition the caller can define a minimum number of servers which should be made available by WLM regardless of whether work is available to execute or not. If the user defines multiple service classes to give the work of the application different service goals, the caller can define that the minimum number of servers is spread across these service classes to ensure that servers are available for all work executed by the application.

The caller must have previously connected to WLM using the IWMCONN service specifying SERVER_MANAGER=YES and SERVER_TYPE=QUEUE. It is recommended to use the IWMSLIM service directly after IWMCONN. If any server uses this service to define limits, the limits apply for all servers of the application environment regardless of whether other servers use the service or not.

If a server defines new limits during execution, WLM attempts to meet the new limit definitions as soon as possible. If the maximum limit for servers is reduced during execution it is not predictable when WLM is able to meet the new maximum definition. This depends highly on the execution time of the running work requests. Therefore changing the limits during execution should be used very carefully and primarily during times of low application utilization.