I agree with samonsey. You should fix this in the staging area when you load the dimension and fact tables.
But Cognos product Transformer wich builds Cognos cubes
fixes this fore you. But that is not always a good thing.
Because then you will build cubes that actually contains data that is not...
Do you not mean that there is a n:n relationship between the fact table and dimensiontable?
In that case you have a select distinct count problem.
And they are hard. Microsoft has function for that.
But it can be slow.
Create a calculated measure called Sales_blablabla with the formula:
ValidMeasure ([Measures].[Sales])
I had the same problem and tried to use the cognos solution
but found this solution instead
hope this will work
Best Regards
Patrik
I found the answere.
Create a calculated measure called mynewmeasurethatworks everywhere with the formula:
ValidMeasure ([Measures].[myoldmeasurethatdonotworkeverywhere])
Hi
I Have several cubes with measures that are combined into
one virtual cube.
The shared dimension are:
Time (Levels: year,quarter,month,day)
CinemaTheater (Levels: Region,city,Cinema,Showroom)
Daytype (onelevel:{weekday,weeken})
Channel (onelevel: {wap,internet,manual})
Movie (Levels...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.