We are creating a Universe in 3.1 againtst a data mart in an Oracle DW. How is a universe setup if the query needs to cross fact tables of different granularity? They have common dimesion tables but the granularity is different from fact table to fact table?
If this can be done, what functionality within Designer allows this?
Any help is appreciated.
Max
MaxVonM 
(BOB member since 2013-02-14)
Hi,
Can you post an example of what you mean? Or provide more details?
Marek Chladny 
(BOB member since 2003-11-27)
This is the best example I could think of quickly without getting into the data
For my example I will have 2 fact tables and 1 common dimension table.
Fact 1 has project member metrics and is only as granular as the project.
Fact 2 has shipping metrics with various dates and is granular down to the project and activities on that project. Multiple activities per project.
The common table is a project dimension table that has various fields relating to the project.
If I queried metrics from both fact tables I think my results would be skewed because of the multiple rows per project on the second fact table.
If this example is not good enough, let me know. This is an oversimplied version of our data, but I think it holds true to how the data is of different granularities between the fact tables.
m
MaxVonM (BOB member since 2013-02-14)
Hi,
It’s called a chasm trap. Having 2 fact tables in a universe is typically a chasm trap.
The solution is to introduce contexts into the universe, having 1 context per one fact table. Check more details in this Designer FAQ:
Marek Chladny (BOB member since 2003-11-27)