cancel
Showing results for 
Search instead for 
Did you mean: 

Summing up key figure in Cube - data load from ODS

prem_kool
Participant
0 Kudos
207

Hello Gurus,

I am doing a data load from ODS to cube.

The records in ODS are at line-item level, and all needs to be summed up at header level. There is only one key-figure. ODS has header and line-item fields.

I am loading only header field data ( and not the item-field data) from ODS to Cube.

I am expecting only-one record in cube ( for all the item-level records) with all the key-figures summed up. But couldn't see that.

Can anyone please explain how to achieve it.

I promise to reward points.

=====

Example to elaborate my point.

In ODS

Header-field item-field quantity

123 301 10

123 302 20

123 303 30

Expected record in Cube

Header-field Quantity

123 60

====================

Regards,

Pramod.

View Entire Topic
prem_kool
Participant
0 Kudos

Hello Oscar and Paolo.

Thanks for the reply. I am using BW 7.0

Paolo suggested:

>>If you don't add item number to cube and put quantity as adition in update rules >>from ODS to cube it works.

I did that still I get 3 records in cube.

Oscar Suggested:

>>What kind of aggregate do you have for your key figure in update rules (update >>or no change)?

It is "summation". And it cannot be changed. (Or at least I do not know how to change it.)

There are other dimensions in the cube - which corresponds to the field(s) in ODS.

But, I just mentioned these two (i.e. header and item-level) for simplicity.

Can you please help?

Thank you.

Pramod.

Former Member
0 Kudos

Hi,

Try this:

Compress and then check the records in the cube.

With rgds,

Anil Kumar Sharma .P

Former Member
0 Kudos

Do you have any other IO diferent from the three specified before and which was different for each item?. Else you will have one aggregated values.

Former Member
0 Kudos

Ok, so you posted an example of what you have in the ODS and what you'd expect in the Cube, can you post an example of what you actually get in the Cube, the 3 rows must be different by something.