Hello All,

Background: I am relatively new to using OLAP (3 months) and have been
working with SQL server for 5+ years now.

I am trying to create a Cube that utilizes the textbook Kimball Bridge
table/MultiValued dimension model:

tblAccount (Fact Table : Unique Key/PK = AccountID)
tbljoinAcctUser (BridgeTable one to many.....AccountID to UserID)
tblUser (Dimension Table: PK= UserID)

In a CUBE, ideally I would like to use AccountID as a measure and
return the appropriate sum of AccountIDs if USER is applied as a
dimension.

I realize this is surely a basic concept but I can find little
information about implementing in ANalysis Services.

Is it feasible to create a CUBE in Analysis Services with a Bridge
table design?
(* I read this article (
http://msdn.microsoft.com/library/de...llcriteria.asp
) and it seems that this concept is not supported in Analysis
Services)

Thanks!
John

I seem to be stating the same thing over, and over again today so I
apologize to the group in advance. You could use a virtual cube to "bring
in" the bridge table but I don't think you want to go there. Use a view on
the backend to make the central fact look single-valued and build the cube
off of the view.

SELECT fact.AccountID,
bridge.UserID,
fact.Measure * bridge.WeightFactor
FROM tblFact fact INNER JOIN tbljoinAcctUser bridge
ON fact.AccountID = bridge.AccountID

"John Townsend" <jt (AT) archeranalytics (DOT) com> wrote

Thanks Martin...much appreciated

"Martin Mason" <martinma (AT) mail (DOT) wt.net> wrote