If you are unable to create a new account, please email support@bspsoftware.com

 

News:

MetaManager - Administrative Tools for IBM Cognos
Pricing starting at $2,100
Download Now    Learn More

Main Menu

What if we don't have any joins in our datasource(s)?

Started by Jaspire, 10 May 2008 07:47:11 AM

Previous topic - Next topic

Jaspire

Hi,
To build a powercube, I want to use each dimension table as a separate iqd datasource and fact table as another iqd datasource without any joins between those datasources. I will just build the dimension map using the columns in each datasource and then use the fact table for measures. Will I get correct results? I saw in documentation somewhere that we should not expect the powerplay transformer to do joins for us. If I don't do joins in the datasources ans use several independant datasources, how will the displayed data be? Will I get correct results? Do I have a chance of getting wrong results? or Is there any other factor that effects that accuracy of data displayed in this case.
I appreciate if you could suggest me even smallest issue related to this.
Thanks,
Jaspire

rockytopmark

Actually, I prefer the method which you will be using... True Dimension and Fact queries, and it will work very well and efficiently.

In Transformer, the relationship between the Dimension and Fact IQDs is made by the Column Names.  So, if your Time Dimensions is built from Year down to Date, and your Date Column is named DateKey for instance, your Fact should also have a DateKey column, of same datatype, which will 'relate' them and give scope in the model.

almeids

100% agree with rockytopmark's comments.
Separate dimensional iqds make for nice modularity and reuse in other models.