Housekeeping for SAP Payroll

Every system creates data that, after a while, you just don't need. And all that data can make things run slower too - so it's good to do some housekeeping now and then. Clear out that old data that you don't need and freshen up the place. And, do it automatically. Here are a few tips for doing that in SAP Payroll:

  1. Close and delete old processes in the HR Process Workbench. If you are using process models (transaction PEST to build, PUST to run) for executing your payroll processes, after a while you can just get rid of the ones that are old. Maybe you keep them around for 60 or 90 days so that yo can go back to previous payroll runs and see who did what, but after that no one really cares and no one needs that data. Keeping around too much can slow your processes too. So - in transaction PUST you can select an old process and close it out (Edit > Process > Close) and then delete it (Edit > Process > Delete). But why go through that trouble? Do the same thing with program H99UCDP0. Make a variant that selects all processes started earlier than 60 days ago by using dynamic date selections, then schedule it to run monthly. Make sure to close and delete the processes.
  2. We always run payroll posting simulations before doing a real posting for payroll, right? Of course you do. And after a while those simulation runs build up and take up space, most notably in table PPOIX. So, you can delete them via transaction PCP0. Or, use program RPCIPQ00 to automaically delete them. Unfortunately there is no dynamic date selection in this program, so you have two options. You can just delete all simulation runs every month or so. That might be inconvenient if payroll is being run and spans a month-end. Or, you could write a 'wrapper' program that selects the posting runs based on some criteria and then executes RPCIPQ00 with that list of runs. This might be a better way to go, particularly if you can't get all payroll countries to agree on how long they need the posting simulations around.
  3. TemSe files don't have to stay around forever - that is why they are TEMporary SEquential files. Whenever a TemSe file is created, it gets stamped with an expiration date. There are usually Basis jobs that delete expired TemSe files, which is good. But what is bad is that a lot of the files created by HR processes don't get tagged with an expiration date. You can fix that by configuring a feature named, appropriately, TEMSE. You configure this feaure as features go, with whatever decision criteria you get from structure PME31, and return the number of days from today that the TemSe object will expire. The tricky part is knowing that 'HR TemSe object identifier' strng to use for a given application. I wish I had good news for a list of all those, but I don't - I usually search and debug the program to find the value I need. If you have a list, please do share.

So those a three things to do to automate some housekeeping for SAP Payroll, Let me know if you have more - because there is always more!

up
205 users have voted.

There are 3 Comments

RPCDCT_DEL_DCT_DATA, that can be used to delete obsolete declustered payroll results..

up
187 users have voted.

Hi Steve, In accordance to the data retention period, what's the best recommendation to auto purge terminated employees (inactive for X years) from EC-Payroll including the payroll results and postings ? The inactive records will be purged from EC, will it then replicate to EC-payroll via PTP?
Is there a write-up or documentation on this ? Thank you.

up
67 users have voted.

Hi Marcus - I haven't seen many companies care so much about purging HR-related data unless they are doing so on an individual basis for GDPR compliance. Report RPUDELPP is a good tool, and you could also use the ILM module on the SAP-side of things. I haven't not seen that purging in EC automatically leads to purging in ECP but it may have been in release notes that I missed. The SAP product pages are a good resource for this (help.sap.com)

up
63 users have voted.

Add new comment

randomness