Specifies that the job is mission-critical and must be processed accordingly.

A mission-critical job gets privileged treatment. Given its deadline and estimated duration, the scheduler:
  • While building the plan, or every time it runs the submit command, calculates the latest start time each of its predecessors can start so that the job successfully meets its deadline. This is called the critical start time. The critical job and every one of its predecessors are assigned a critical start time.

    The entire set of predecessors to the critical job is referred to as the critical network of the job.

  • While running the plan, dynamically recalculates the critical start times within the critical network.

    When a predecessor risks compromising the timely completion of the critical job, it is promoted; that is, using different operating system mechanisms, such as implementing the nice command on UNIX or changing the priority level on Windows, it is assigned additional resources and its submission is prioritized with respect to other jobs that are out of the critical network. This action is recurrently run on any predecessor within the critical network and, if necessary, on the critical job as long as there is a risk that this job becomes late.

Important: Critical jobs, included in the plan because associated to a run cycle, must have a deadline specified at job, job stream or run cycle level. Whereas, critical jobs submitted in plan on request might not have a specified deadline, and, in this case, the global option deadlineOffset is used.