0

Normally dimension values in a cube are distinct and selecting multiple values means: "give me all rows with either of the values (union)". Instead of this I want the intersection, so only the rows which have all values selected. How do I model this in microsoft analysis services?

So example 1 would be a dimension Store, with a value "store A" linking to multiple related values T1 in a different dimension, and "store B" linking to T1 and T2 and lastly "store C" linking to no values.

What I'm looking for is that selecting T1 gives me store A and B, and selecting T1 and T2 only gives me B, not A.

Anders Rune Jensen
  • 3,758
  • 2
  • 42
  • 53
  • 1
    Can you illuminate this with an example? How can a fact have two values for the same dimension? Are you looking for a CROSSJOIN? – Tab Alleman Feb 17 '16 at 14:49
  • Updated with example, I started out with something like this: https://stackoverflow.com/questions/10385513/ssas-one-to-many-dimensional-relationship. It works but only as union, not as intersection. – Anders Rune Jensen Feb 18 '16 at 07:28
  • Your example sounds more OLTP than OLAP. Can you re-state your example in terms of facts and dimensions? What is value T1? A dimension value? A measure value? Are you sure you should be using a cube at all? Your goal doesn't sound suited to SSAS. – Tab Alleman Feb 18 '16 at 14:04
  • @TabAlleman Thanks for the clarification notes, see my revised question and the link. That is how I have modelled the store can have one to many values for T. – Anders Rune Jensen Feb 19 '16 at 08:18
  • You don't link dimensions to other dimensions in SSAS. You either combine them into one hierarchical dimension, or you let them link individually to the fact table. – Tab Alleman Feb 19 '16 at 13:24

0 Answers0