Forum .LRN Q&A: Re: .LRN v2 Performance

Collapse
4: Re: .LRN v2 Performance (response to 1)
Posted by Tracy Adams on 09/10/04 06:26 PM
One thing just fixed -  The Oracle query for sending bulk mail to people who HAVE not taken the survey for large groups was very slow.  I rewrote the query and submitted. I'm committing now.

We are also having issues with sending lots of emails (notificaiton and bulk mail) which Andrew is working on.

It's uncanny, the Oracle on our new box is coming up with very different query plans that we are used to seeing, particually with IN clauses with subqueries.  The same query will run very differently on our old production box than on our new dev and production boxes.

Collapse
7: Re: .LRN v2 Performance (response to 4)
Posted by Dirk Gomez on 09/11/04 09:30 AM
Tracy, the same Oracle versions? Vastly different init.ora parameters? Up2date statistics?

Actually the CBO should be deterministic. Kinda...it has earned its crystal ball repuation because it is different from minor version to minor version.

You also don't say whether the queries on your new system perform worse?

Alas, stored outlines (asktom.oracle.com and the Kyte book have a lot on it) are not an option because the in subclauses are not using bind variables and there is simply too many different queries executed even with cursor_sharing=force.

(Have you set cursor_sharing = force ?)