Multiproviders - identify characteristics and select key figures

[Former Member]

Hi,

I'd appreciate some clarification on the identify characteristics and select key figures steps of creating a multiprovider. What is the purpose of these steps? I didn't find anything helpful in this forum or in the SAP Help documentation. If there are 2 cubes in the MP, and both have the same characteristic like 0salesorg, am I supposed to select 0salesorg from each of the cubes or just 1 of them? What about a situation when 1 cube has 0division and the other has 0division and a navigational attribute 0division? Am I supposed to select 0division from the first cube and either the 0division characteristic or 0division navigational attribute from the other cube? Regarding key figures, if both cubes have the same KF, am I supposed to select only 1 of them to avoid doubling up of an amount, for example?

Thanks,

John

[Former Member]

Thank you all for your answers to my questions. I've awarded points. John Burk

[Former Member]

The 'iDENTIFY' helps us to decide fro whch cube , we need a field if it exists in both of them.

IF u think it is available in both and same in both, you can also choose to identify from both.

In case of Keyfigure, also it works similary and the values will not duuble up if selected from both cubes.

Each objects needs to be atleast identified from atleast one infoprovider.

[Former Member]

Hi there,

Regarding the characteristics and key figures assignment, the purpose of the MP is to unify the data between its InfoProviders.

So, if you have 0amount in cube1 and 0amount in cube2, if you don't identify in the MP that 0amount comes from cube1 and cube2, you won't see 0amount in the MP. If you assign 0amount only from cube1, then you'll only see values of 0amount from cube1, no data regarding of cube2 with 0amount will be displayed on MP, and vice-versa

This stands either for key figures and for characteristics.

Now, regarding the 0salesorg__0division:

Imagine you have in your cube1 0salesorg, 0divison and also 0salesorg__0division. If you went to see the data from the cube1 regarding 0division and 0salesorg__0division it would be probably the same, so here, matching in a MP either 0division or 0salesorg__0division should also be the same. But you could sometimes have in cube1 0division char without values, that were already loaded on the master data attributes of salesorg (0salesorg__0division), and therefore you would have inconsistency in cube1 between 0salesorg__0division and 0division, so matching 0salesorg__0division of a cube with 0disvision of another could lead to incorrect practice.

The best approach is, if you have both 0salesorg__0division and 0division of a cube to match with 0division of another cube, always use the characteristic itself (0division). The same goes for the nav. attributes: ie., if you have 0division and 0salesorg__0division in cube1 to match with 0salesorg__0division of cube2, always use the nav. attribute.

Hope this helps,

Regards,

Diogo.