TM1 Forum

Hi,

Based on my scenario, the current and future weeks / months data is not consolidating properly (When I trace the rules, it looks like "not fed"). Only few are fed in the feeders for future months. FYI, Feeders are conditional.

To achieve this, I have created a process with "CubeProcessFeeders ('CubeName')" which took 2 hours 55 minutes for execution and got the result (But, I doubt that if the data is updated further then the feeders "not fed" issue will be repeated). Prior to my process execution, the RAM utilization was 65 to 74 GB (Normal) and now it is 92+ GB which is unusual (Cube file size, persistent feeder file size remain same).

Mean while, I am trying to include the static parameter "ForceReevaluationOfFeedersForFedCellsOnDataChange=T" in configuration file (My assumption is that, the data which will be updated in feature shall have the feeders fed. But the old/existing data may not have any effect).

Can any one help me on this.
FYI, server details
RAM : 136 GB
TM1 : IBM Cognos Express v10.1

So you have a cube in which a rule save / feeder processing takes around 3 hours and consumes about 30 GB of memory for feeders?

Off the cuff I'd say what you need is for a competent TM1 developer to health-check the model. I can't give you anything beyond that advice as you haven't given the faintest clue as to the dimensionality of the cubes and the code of the relevant rules and feeders and without this information no one else will be able to help either. If you need help then read the request for assistance guidelines and at least try to make it easier for others to help you.