Hi,
I have a strange situation that I'm hoping someone can help with.
We have a dimension that contains all employees in a company and the teams that they are in. An employee can appear in more than one team, hence a many-to-many dimension.
When we process the dimension we observe that some of the team members are missing from some of the teams. We have concured that when we process, the correct queries are being issues against our underlying star schema. However, some dimension members turn out to be missing after processing.
I know this is a very vague description of the problem, and I'm usually the first person to criticise in that situation, but I was hoping someone might have pearls of wisdom in regard to many-to-many dimensions.
Is there any gotchas that I need to be aware of
Anything I need to setup that isn't obvious
etc...
Thanks in advance!
-Jamie
Some extra info here...
I have since discovered that in the Team I am currently using for investigation, the only members that show up are the members that are ONLY in that team. The members that are in another team show up elsewhere.
This leads me to think that we haven't configured our many-to-many dim successfully.
Missing dimension members in many-to-many
A. Jason
OK, i got to the bottom of this. We were trying to build a hierarchy out of the 2 dimensions which was the wrong thing to do. We needed to treat them as seperate dimensions.
We still needed the many-to-many hierarchy though so we did the join of the 2 tables in the DSV and then used the Cardinality property to define the many-to-many relationship between the hierarchy levels.
That's a very cool new feature of SSAS I reckon. I'm very impressed.
-Jamie