Using multiple cubes within a single report 10.2.1

[gosoccer]

Friends,
We are in the need of using multiple cubes within the same Report Author 10.2.1 Report using Dimesional H.
Or any cube implementation that achieves the same!

We have noticed a much faster report when the cube is smaller in size!
I would very much appreciate your recommendation !!

[MFGF]

Friends,
We are in the need of using multiple cubes within the same Report Author 10.2.1 Report using Dimesional H.
Or any cube implementation that achieves the same!

We have noticed a much faster report when the cube is smaller in size!
I would very much appreciate your recommendation !!

It's possible to create packages that contain multiple cubes, but whether this delivers the results you are looking for depends on what your requirements are. Data from two cubes cannot easily be used in the same reporting object (list/crosstab/chart etc) since there is no concept of being able to "join" cubes. You could nest a crosstab from one cube within a list from another cube (for example) and link them using a master/detail relationship (based on matching captions) but as you can imagine, it makes the authoring process much more cumbersome.

If you are having performance issues, you could consider:

1. Reviewing the cube partitioning strategy to see if access times can be improved
2. Reviewing the suitability of cubes to answer your reporting requirements. Cubes are not good at delivering vast list-style reports.
3. Assessing whether you are using too much data for a PowerCube to be the appropriate OLAP technology
4. Investigating alternatives such as Dynamic Cubes where you can leverage in-database aggregates and in-memory aggregates

Cheers!

MF.

[gosoccer]

 :)WHOOM! interesting.
What do you mean by Cube Partitioning Strategy? Could you elaborate on that a little bit.
thank you so much!!

[MFGF]

:)WHOOM! interesting.
What do you mean by Cube Partitioning Strategy? Could you elaborate on that a little bit.
thank you so much!!

Hi,

It's an involved topic - so rather than me trying to provide a hamfisted attempt at explaining, take a look at the Transformer User Guide, Chapter 8 > Maintaining Models and Cubes > Choosing a Partitioning Strategy

http://public.dhe.ibm.com/software/data/cognos/documentation/docs/en/10.2.1/ug_cogtr.pdf

Cheers!

MF.

[gosoccer]

OK. Great!! I'll review it. We are using Auto-Partitioning but the document discusses other steps I can take to make
things optimized like Consolidation.

Thx