Extensive memory consumption using PullUp-Functions

Hello and a happy new year,

we do have the problem of memory allocation while using several Pull-Up Aggregations. Is there a way to release the memory frequently?

Our set up is Hybrid (IBC + Leave your data in place).

We have 90mio

Activities@8.5mio

Cases in one Model and 260mio Activities@24mio Cases in a second Model. The memory consumption is in idle state at 24 GB and moves up to 64 GB when using PU-Functions. This will quite often let the engine crash and models could not be re-loaded even from cache until the memory garbage is cleared.

Thanks for any suggestion.

Regards

Holger

Page 1 / 1

Hi @holger.meiss,

normally Pull Up Function are more performant than other PQL queries like a CASE WHEN. Also the amount of data should be fine. It is recommended to not exceed 800 Million activities per data model. You are far below. Could you please raise a Servicedesk Ticket and check with our team what the actual issue is?

If you have further questions please feel free to reach out!

Best,

Justin

@holger.meiss are you using FILTER_TO_NULL in you PU Functions? This can be very expensive computational wise.

Hi Holger,

do you have filters, CASE WHEN or aggregations that you could move it to SQL? This would reduce the calculation required during the interactions in the Analytics.

Best,

Gabriel

Reply

ANY QUESTION?

MORE CELONIS

CHECK US IN

Reply

Sign up

Login with SSO

Login to the community

Login with SSO

Scanning file for viruses.

This file cannot be downloaded

ANY QUESTION?

MORE CELONIS

CHECK US IN