By default there are some agents that does the house keeping of the cache tables and some of the system tables. You need to ensure that these agents are up and running and also appropriate retention period is configured for them.
Other than the above, housekeeping of the application data or work objects tables should be done by configuring the Purge & Archiving process. Please note that you can tidy up the rules tables by deleting the unwanted rules/rulesets, skimming the rulesets.
Below are some of the tables that has large data at your end.
PR_index_reference:- Stores instances of the Index-Reference class, which maintains references to rules used within other rules.
Spr_sys_updatescache:- System pulse refers this table. The tables stores instances of the System-Updates-Cache class (rule updates to be distributed to other nodes). Contains one row for each rules change implemented since the most recent pulse. This table is scanned when rules are updated and added to the cache, and is emptied when cache processing completes. Check the frequency of system pulse.
Pr_perf_stats :- Stores instances of the Log-Usage class, used in analysing system performance. Again this table data can be controled by configuring the retention period.
Thanks for your replies. Did not reply earlier as I wasn't aware of your replies as no automatic notification was sent to me by PDN.
I can also see from the excel list size of the tables which are big however what I am looking for is definitive details of which agents and what configuration (retention etc.) I can change to reduce the size. Do we know which specific configuration I can change to reduce the sizes for each of the tables?