Hello,

My name is Soumitra Dutta oxford based entrepreneur , I'm working with very large datasets in IBM analytics tools and running into performance challenges. What approaches or best practices have you found helpful? I'd appreciate any suggestions from the group.

Regards,

Soumitra Dutta

------------------------------
Soumitra Dutta
------------------------------

Hi Soumitra, 

the best you can do is making tables small (e.g. by additional aggregated tables or materialized views). If that is not possible you might have a look at Cognos Analytics data sets. Here is a nice article about how to build (aggregated) data marts with them: 

https://community.ibm.com/community/user/blogs/andrew-copeland/2022/10/13/creating-an-optimised-data-architecture-using-data

Please keep in mind that if you join data sets with other sources you will pull raw data from these other sources to be process on the Cognos server instead of the database server. If this is needed you can try join optimization in data modules. 

Best regards,

Philipp

------------------------------
Philipp Hornung
Product Owner & BI Manager
Techniker Krankenkasse
Hamburg Germany
#IBMChampion
------------------------------

Hi all. 

For large datasets, I also think it's a good idea to store everything in DB2:

https://www.ibm.com/docs/en/cognos-analytics/12.1.x?topic=dsoufds-storing-data-from-uploaded-files-data-sets-in-db2

Jens 

------------------------------
Cognos Tipps: https://www.cognoise.de
Lernvideos: https://apparo.university (also in english - please PM)

Jens Bäumler
Cognos Analytics, Planning Analytics and watsonx BI
Apparo Group, Germany
www.apparo.de
------------------------------