Question:
Hi All,
In our cube we do have Material / plant.
As per our query we need to have both of them in Aggregates.
But our performences is going down for material.
I want to have another Agg for Plant Only.
How about
Answer:
Well,
I am not 100% sure, but for me this is highly NOT recommended to have more than one aggregate on a cube ; because this would duplicate the data and then instead to have the double amount in case of only one aggregate (per plant, for instance), there is a risk that you have three times the same amount if you had the material (for instance) in this aggregate. Before even considering to start with this, you should maybe ask a professional that attended the Performance training at SAP (BW360), or maybe an Oracle specialist can have a look, provided that you are running on Oracle, of course. Good luck.
Answer:
I'm quite OK with your answer, ChC, except if you make some calculations in your queries, creating a exception aggregation which divides the result by the number of potential aggregates, you should be able to report the correct value, indeed. the check is also possible by counting the number of documents, and dividing it by the number of tuples in the fact tables, and applying the resulting calculation to all the key figures. Depending of course on the fact that you have key figures with a standard aggregation equal to summation. If you manage non dimentional key figures in the key figures area (some dates), I guess you can also apply some shifting on the results, or maybe enter multiple fiscal variants, dummy ones (for instance 01, 02, 03 .... one for each aggregate defined) which will enable you to create a dynamic selection of the dates.
hope this help.
Good luck.
_________________
When my car stops, I first look at the gasoline's level before dismounting the engine.
Tuly Idiot's fan club active member.
Answer:
Thanks CHC,Fred.
But have column filter by day and month.
my aggregates on month , but its not looking at Agg Cube.
I knew since day is not in agg cube its need to read Base Cube.
But its not reading Agg Cube for Month columns ,but still reading the Base cube.
Its not driven by day..? Whats the hell in SAP
Thanks
Raj
Answer:
Maybe Optimisation daemon decided to do so. Since it has to read the day by day, decided to add the day data instead of reading another table.
_________________
When my car stops, I first look at the gasoline's level before dismounting the engine.
Tuly Idiot's fan club active member.
Answer:
If your query is ran about history data, you should be careful about year 2004. Effectively, 2004 had one day more than 2003, for instance, due to Feb29, so this part would take longer obviously. But ok, this remark is not valid if you are only dealing with data of 2005.
Ch