Hi,

I'm just looking for some ways to improve performance of queries that touch data sets.

I'm curious about one advanced setting of query service: qs.queryExecution.flintServer.sparkThreads

https://www.ibm.com/docs/en/cognos-analytics/11.2.0?topic=reference-query-service-advanced-settings

qs.queryExecution.flintServer.sparkThreads

Specifies the maximum number of threads that the Compute service can use to service queries.

The specified value must be a positive integer greater than 1.

Data type:

Positive integer

Documentation is not clear about what is does and the default number of threads. Would be 1? Or another value?

Does anyone has changed this property? Any good results?

Thanks!

------------------------------
JEAM COELHO
Cognos Solution Architect

LinkedIn: https://www.linkedin.com/in/jeamcoelho/
------------------------------

#CognosAnalyticswithWatson

I got an answer from support

Based on the information I found on qs.queryExecution.flintServer.sparkThread Query Service advanced setting, this is an automatically calculated parameter and it should not be changed under normal circumstances.

------------------------------
JEAM COELHO
Cognos Solution Architect

LinkedIn: https://www.linkedin.com/in/jeamcoelho/
------------------------------

It pertains to the number of concurrent API requests between an instance of Dynamic Query and Compute service.

There should never be a reason to change the default.

It will not change the execution time of a query processed by the Compute service.

------------------------------
NIGEL CAMPBELL
------------------------------