Author Topic: Virtual Cube Behaviour in Analysis Studio  (Read 421 times)

Offline cognos05

  • Statesman
  • ******
  • Join Date: Apr 2014
  • Posts: 756
  • Forum Citizenship: +1/-1
Virtual Cube Behaviour in Analysis Studio
« on: 13 Nov 2020 01:40:04 pm »
Hi,

I have a cube A
whic has customer, product, sales etc..

Second cube b is
produt and budget

I created a virtual cube with cube a and cube b

Now i have product in my rows and my measures in column as sales and budget

Now i wanted to see my sales for only certain egions in customer ,

So i add a slicer on customer dimension , once this added my budget becomes zero since its not realted to customer dimension .

In transformer i think the values will show even if the dimension is not related , same budget value for all customer .

so can i not see the budget and sales if a non conformed dimension is present in a filter .

Offline bus_pass_man

  • Statesman
  • ******
  • Join Date: May 2008
  • Posts: 480
  • Forum Citizenship: +45/-0
Re: Virtual Cube Behaviour in Analysis Studio
« Reply #1 on: 15 Nov 2020 11:48:57 am »
The fact that a query involving a measure and an object from a dimension which is not conformed to the measure returns no results is a good thing.

Transformer allows for the arbitrary allocation of a measure to dimensions.
https://www.ibm.com/support/knowledgecenter/SSEP7J_11.1.0/com.ibm.swg.ba.cognos.ug.cogtr.doc/c_allocatingmeasures.html

There is the ability to allocate measure values below the fact grain in dynamic cubes.

https://www.ibm.com/support/knowledgecenter/SSEP7J_11.1.0/com.ibm.swg.ba.cognos.ug_cog_rlp.doc/c_cog_rlp_calc_mem_uses.html

In both cases, the allocation may or may not be as you might want it. Worse, the consumers of the cube might not know about the allocation or understand its significance, and make decisions based on the possibly invalid data.

To have full control over the determination of what values are assigned to each member of the levels in the dimension where you want it would be better that, if you have a situation in which you want to have a plan fact for a certain dimension, then the appropriate planners ought to determine those values and have them added to your data warehouse, where they can be built into the cube.




 











Offline cognos05

  • Statesman
  • ******
  • Join Date: Apr 2014
  • Posts: 756
  • Forum Citizenship: +1/-1
Re: Virtual Cube Behaviour in Analysis Studio
« Reply #2 on: 16 Nov 2020 09:12:02 am »
Thanks for the link ,

I have thsi calculate measure

value(completeTuple(currentMember(Fiscal Year_H),All,All,All,All,All,All,CurrentMember(Items_H),Budget))

The issue is i have 2 hierarchies on time ,so how should i make sure it works for both hierarchies. Current Members allows only 1 hierarchy .