Understanding Job History

The job record retention period (TTLCompletedJobs) determines how long to store data for the following records:

  • Data about completed jobs (Finished, Failed, or Canceled) in the HPCScheduler database. The information and metrics in the HPCScheduler database are displayed in the Node Management and Job Management views in HPC Cluster Manager.

  • Diagnostic test results and data in the HPCDiagnostics database.

  • Messages for completed durable sessions that are stored by the broker node using MSMQ. By default, the MSMQ storage limit is 1 GB. Depending on the message size and the number of durable sessions running on your cluster, you can modify the record retention period or increase the MSMQ storage limit.

Jobs that are in the Configuring state are not deleted from the database. The jobs must be canceled by the job owner or a cluster administrator (or completed in some other way) and then they will be deleted according to the job history policy.

For information about how to change the configuration options, see Configure the HPC Job Scheduler Service.

Note

The HPCReporting database stores information about jobs and nodes for reporting. For more information about the HPCReporting record retention parameters and how to view reporting data, see Charts and Reports: HPC Cluster Manager.

Additional references