Workload service assurance

Basic concepts about critical jobs and the critical path, which are fundamental to take advantage of this feature.

Workload service assurance is an optional feature that allows you to identify critical jobs and to ensure that they are processed in a timely manner.

When the workload service assurance feature is enabled, you can indicate that a job is critical and define a deadline by which it must be completed when you add the job to a job stream. Defining a critical job and deadline triggers the calculation of timings for all jobs that make up the critical network. The critical network includes the critical job itself and any predecessors that are defined for the critical job. When changes that have an impact on timings are made to the critical network, for example addition or removal of jobs or follows dependencies, the critical start times are automatically recalculated.

The critical network is constantly monitored to ensure that the critical job deadline can be met. When a critical network job completes, timings of jobs that follow it are recalculated to take account of the actual duration of the job. The system also acts automatically to remedy delays by prioritizing jobs that are actually or potentially putting the target deadline at risk. Some conditions that cause delays might require your intervention. A series of specialized critical job views, available on the Dynamic Workload Console, allow you to monitor critical jobs, display their predecessors and the critical paths associated with them, identify jobs that are causing problems, and drill down to identify and remedy problems.

Dynamic critical path

If a job is critical and must complete by the deadline set on the database you can mark it as a critical job thus specifying that it must be considered as the target of a critical path. The critical path consists of the critical job predecessors with the least slack time. In a critical job predecessor path, the slack time is the amount of time the predecessor processing can be delayed without exceeding the critical job deadline. It is the spare time calculated using the deadline, scheduled start, and duration settings of predecessors jobs. The calculation of critical path is performed dynamically. In this way, during daily planning processing, a critical path including the internal and external predecessors of the critical job is calculated, and a table of predecessors is cached (in the local memory for z/OS and on the master domain manager for distributed systems). Every time a predecessor of the critical job starts delaying, the scheduler dynamically recalculates the critical path, to check whether a new path, involving different jobs, became more critical than the path calculated at daily planning phase.

You can launch a query for all the jobs included in a critical path by clicking Critical Path in the panels that show the results of monitor jobs tasks.

As well as jobs included in the critical path job list, there are other lists of jobs that might be important to manage to ensure that your critical job does not fail.

Hot list

The hot list contains a subset of critical predecessors that can cause a delay to the critical job because they are states such as error, late, fence (for distributed systems only), suppressed (for distributed systems only), or long duration. If these jobs do not complete successfully on time, they prevent the critical job from completing on time. Using the hot list view, you can quickly see which jobs need you to take appropriate recovery actions. Jobs included in the hot list are not necessarily also included in the critical path.

You can launch a query for all the jobs in the hot list by clicking Hot List in the panels that show the results of monitor critical jobs tasks.