Rearchitect the colgroup processing to be more granular using multiple sorts as opposed to a single sort or more intelligently avoid the colgroup sort altogether (when possible)

Looking at the job output you sent in, we can see that there are 18 colgroups specified.

Based on the current design, we have a single sort for colgroup(s).

There are 184310123 rows unloaded

The no. of the rows passed to the single

dstats sort pipe is equal to 18 x no. of records inside the tablespace.

This may take a long time to do the sort.

There is no easy way to fix this scenario based on the current design.

You can choose to reduce the amount of data being collected and processed by

specifying less the number of colgroups.

As a future enhancement we need to rearchitect the colgroup processing to be more granular

using multiple sorts as opposed to a single sort or more intelligently avoid the

colgroup sort altogether (when possible). You can open an AHA request on this.

Needed By

Quarter

Post comment

Guest

Reply
| May 6, 2022

Thank you for submitting this enhancement request. While we view all the enhancement requests we receive as valuable and we would like to implement all of them, in an effort to have maximum transparency we are closing the ones that don’t currently appear on our two year product roadmap.

We have chosen to decline this request at this time for the following reasons:
The suggested re-architecture of the colgroup processing is a large development effort and would require delivery in a release boundary with significant changes. Instead, we are working towards a long-term solution to improve stats profile processing to reduce the necessity of generating a large number of colgroups.

0 reply Hide replies

By clicking the "Post Comment" or "Submit Idea" button, you are agreeing to the IBM Ideas Portal Terms of Use.
Do not place IBM confidential, company confidential, or personal information into any field.

Please enter your email address

RELATED IDEAS

Rearchitect the colgroup processing to be more granular using multiple sorts as opposed to a single sort or more intelligently avoid the colgroup sort altogether (when possible)