Case that runs slow:
We have a big database and the cube has been in a 9.3.1 environment.
This is a 10 dimensional cube with 2 dense dimensions.
We have a script running in more than 2 days, which has commands :
SET CALCPARALLEL 4;
SET CALCTASKDIMS 3;
SET UPDATECALC OFF;
SET AGGMISSG ON;
SET FRMLBOTTOMUP ON;
FIX("21000", "22600", @REMOVE(@DESCENDANTS("All Products"), @DESCENDANTS("Reclaimed Rubber")), @REMOVE(@DESCENDANTS("Total Customer"),@DESCENDANTS("MICHELIN NORTH AMERICA, INC.")))
SET CREATENONMISSINGBLK ON;
IDR = NA_Partner->NA_Misc->NA_Product * ("Sales Volume"->IDR->Marketing / ("Sales Volume"->IDR->Marketing->"All Products"->"Total Customer" - "Sales Volume"->IDR->Marketing->"All Products"->"MICHELIN NORTH AMERICA, INC." - "Sales Volume"->IDR->"21000"->"Reclaimed Rubber"->"C-0005-01"));
SET CREATENONMISSINGBLK OFF;
ENDFIX
and more aggregate commands similarly ....
----------------------------
Method to Increase Calculation speed:
1.When the intelligent calculation is turned off and CREATENONMISSINGBLK is ON, within the scope of the calculation script, all blocks are calculated, regardless if they are marked clean or dirty.
2.Cross dimension operators can be reduced, include more items inside FIX instead of using "->"
