COGNOiSe.com - The IBM Cognos Community

IBM Cognos Analytics Platform => Cognos Analytics => Dynamic Cubes => Topic started by: mkarthykeyan on 19 Apr 2021 04:02:10 AM

Title: Cartesian between two dimensions
Post by: mkarthykeyan on 19 Apr 2021 04:02:10 AM
I have two dimensions and FACT in a dynamic Cube.

FACT
a   1   xyx   100
a   1   hju   50
b   2   xyx   100
b   2   hju   50
c   3   xyx   100
c   3   hju   50




Dim1 :

Product
a
b
c

Dim2

Department
1
2
3

what im getting

a   1   150
a   2   
a   3   
b   1   
b   2   150
b   3   
c   1   
c   2   
c   3   150

what i expect

a   1   150
b   2   150
c   3   150

Need help how to avoid duplicate or Cartesian ...
Title: Re: Cartesian between two dimensions
Post by: bus_pass_man on 19 Apr 2021 07:03:47 AM
I don't get that. 

A little bit more information about how the dimensions and the cube have been modelled and the structure of the data base would be helpful.

I'm assuming that these are separate dimension tables and fact tables.

How have you modelled the relationships between the dimensions and the fact? 

Are they all 1.n?

What are the relationships between objects in the dimensions?


Are the xyx and hju stuff part of another dimension table?

What are you doing to create your report?  Are you using levels, sets of things? Members?
Title: Re: Cartesian between two dimensions
Post by: mkarthykeyan on 19 Apr 2021 08:16:17 AM
I'm new ... still learning.. i try to answer..

yes separate FACT and DIM tables..
In Cube designer we don't have option like 1:n or something like that .. rather than just we have operator like =,>,< etc..Established the relationship using = operator ... XYZ stuff sit in the same FACT table... I'm creating report using level