Assign base priority when you create a task. The values are “high,” “medium,” and “low.” There is a run queue for each priority for each engine, and the global run queue also has a queue for each priority.
When a thread pool looks for a task to run, it first checks its own high priority run queue, then the high priority global run queue, then its own medium priority run queue, and so on. The result is, runnable tasks in the high priority run queues are scheduled onto thread pools more quickly than tasks in the other queues.
The scheduler search space refers to where engine schedulers look for work (run queue checking):
In process mode, the scheduler search space is server-wide, meaning all engines share a global run queue. Engines check the run queues of all other engines.
In threaded mode, the scheduler search space is specific to the thread pool. Each engine thread pool has its own global queue, and the engines within that pool look for tasks associated only with that pool.
During execution, Adaptive Server can temporarily change a task’s priority if necessary. A task’s priority can be greater than or equal to, but never lower than, its base priority.
When you create a user-defined execution class, you can assign the values high, medium, or low to the task.
