We have multiple cubes in our organization (data is divided by month).
I'd like to create a virtual cube to query across all the cubes but my
problem is that the measure names from cube to cube change, e.g.
volume-oct in one and volume-nov in another.

Is there some way to alias the measure names so I can have a single
virtual measure, e.g. volume?

Thanks,
Jay

Sure, just create a calc member equal to the member.

public @ the domain below
www.tomchester.net

"Jay" <googlegroups (AT) jayritchie (DOT) com> wrote

Tom,

Thanks for the reply. I've been to your site after I came across your
Time Dimension article. I actually didn't do the calculated member
because I erronously thought it wouldn't honor the time dimension, but
apparently it does, so that works. Thanks.

However, along the same vein, not only do our cubes change the name of
the measures from cube to cube, but one of the dimensions, e.g.
customers_dec and customers_jan. So if I return my calulated member on
columns and the dimensions on rows, I only get half the value of my
calculated member for each row.

How do I create a dimension (or calculated member) that encompasses
all the members of multiple dimensions, i.e. a union of dimensions,
that could be returned as a row in a MDX statement.

Thanks again for your help.
-Jay A. Ritchie

"Tom Chester" <publicNOSPAM (AT) tomchester (DOT) net> wrote

I assume there is a good reason why you don't have a single Customers
dimension...?

In your scenario how are the customer dimensions sliced?

public @ the domain below
www.tomchester.net

"Jay" <googlegroups (AT) jayritchie (DOT) com> wrote

Hmmm... good reason? I don't know, I wasn't here when they did this.
The customer dimension is never really sliced, except to perhaps
select a single customer. There's a hierarchy in the customer
dimension, so there are a couple other ways to slice it, but nothing
that would explain the need to create a new dimension each month. It
could be that the hierarchy of customers can change from month to
month, but we ususally roll things up based on how they rollup today,
so the current customer dimension is the one I would want to use in
any scenario, regardless if the cube I'm getting the measure from has
it's own customer dimension.


"Tom Chester" <publicNOSPAM (AT) tomchester (DOT) net> wrote

Jay, there's no way to union the dims together. What you can do is write
conditional formulas that sum your measures across all customer dimensions
(if not sliced); it's pretty gnarly though, espeicially since you have N
Customer dimensions and presumably N grows over time. Here it is half-baked:

IIF (Cust1.CurrentMember is Cust1.[All Cust1],
(MyMeasure, Cust1.[All Cust1]) + (MyMeasure, Cust2.[All Cust2]),
MyMeasure )

public @ the domain below
www.tomchester.net

"Jay" <googlegroups (AT) jayritchie (DOT) com> wrote