Hello
I need your experience in (hopefully) resolving a performance problem while extraction data from CO-PA during the intiali phase.
Our CO-PA is pretty huge and we decided to start initializing the data with smaller packages. These two extractions (1st package: July2007 - September2007; 2nd package: October2007-Januray 2008) lasted not to long and we were very happy about the good performance of data-extraction (before that we had to create the indicies stated on sap note 392635, 210219). So we decided to get a "booking free" time from the users to run the last and outstanding initializing run of CO-PA with period Feb2008 to Dez2100.
Between the last sucessfull initial package and the new run there were two weeks in between and nothing was changed on the system.
The last inital run was not sucsessful because all of the sudden the extraction time rised and rised and at the end there was "no" activity at all and we had to stop. Since then we are not able to locate the problem and what could be the reason for this behaviour. The indizies are still there but in the meantime we are even able to create the first inital loads a second time with good performance. Traces made during the bad extratcion showed that the range for the object-number block (CE4*) is 10.000. The start of the extraction is fine but then after reading some blocks the performance decreases to the behaviour described above.
We run our BI and R3 with MaxDB.
I would appreciate your help our thoughts with this so that we can get a little further.
Bye Markus