How a job queue works

Job queues are allocated by a subsystem via the job queue entry. Jobs can be placed on a job queue even if the subsystem has not been started. When the subsystem is started, it processes the jobs on the queue.

The subsystem description specifies the maximum number of jobs (batch or interactive) that can be active at the same time. The number of jobs that can be active from any job queue is specified in the job queue entry.

Not all jobs on a job queue are necessarily available for processing when the subsystem is started. Scheduled jobs can be placed on the job queue. Jobs can be held on a queue until the system operator releases them. If the subsystem is ended before all of the jobs are processed, the jobs remain on the queue until the subsystem is started again, until moved by the system operator to another job queue, until deleted by the system operator, or until another subsystem allocates the same job queue.

More than one subsystem description can refer to the same job queue, but only one active subsystem at a time can use the job queue as a source of batch jobs. Therefore, if a subsystem ends and jobs are still on the job queue, another subsystem referring to that job queue can be started to process the jobs. If another subsystem is already started and is waiting for the same job queue, the subsystem automatically allocates the job queue when it becomes available.