How are jobs scheduled?
Job scheduling is the mechanism which selects jobs waiting in the queue to be dispatched and executed on the cluster. Scheduling jobs effectively in a shared environment essentially comes down to satisfying the demands of a wide range of users with different job characteristics (cpu count, runtime duration, etc), with the goal of providing fair access to the system. In practice, if there are enough free processors to run your job, and no else has any jobs queued, then you should expect your jobs to start immediately. Once there are more jobs queued than available resources, the scheduler will order jobs waiting to run based on a “fairshare” priority basis. (Note that for jobs submitted to the “test” queue, this does not apply, since they preempt running jobs and do not wait in the queue.) Fairshare is a measure of recent resource usage. All user groups are ranked into 5 priority levels, where the most heavy users are ranked in priority level 1, and the least heavy users are ranked in p
