Cleaning collected Performance Analytics data
Performance Analytics scores and snapshots may grow over time and should be routinely cleaned to ensure optimal performance and accurate data.
Performance Analytics uses a scheduled job to remove old scores and snapshots.
The Clean PA collections scheduled job is active by default and has no impact on performance. By default, the job runs daily so it only has to delete a small amount of data.
Note:
The dictionary attributes nibble_size and
nibble_sleep affect the behavior of the collection cleaner job if the attributes are defined for one of the Score (pa_scores, pa_scores_l1, pa_scores_l2) tables or the Snapshot (pa_snapshots) table. For more information, see Altering tables and fields using dictionary attributes.
Modify the Clean PA collections job
Modify the scheduled job to configure when Performance Analytics scores and snapshots are cleaned. The scheduled cleanup job should not run while a data collection job is running.
Before you begin
About this task
By default, the Clean PA collections job runs at 05:00 which is appropriate when using the default data collection jobs. If you create additional data collection jobs, you may need to change the start time of the Clean PA collections job.