Page 1 of 1

Way to identify multi-row fetch candidates?

PostPosted: Fri Sep 28, 2012 8:56 pm
by Jonathan Jeuck
Hi,

If you wanted to identify the DB2 programs running in z/OS that would most benefit from implementing multi-row fetch, how would you do it? It seems to me that you'd want to run a report that identifies either the program or the plan, and allows you to sort by "total rows fetched over x days", "fetches per open cursor", and "rows per fetch". How would you go about creating a report that shows this for all jobs/programs over the last 30 days? Would you need a third party performance monitor? Are there some simple ways to run a report with this information within DB2 for z/OS?

Thanks all,
Jon