|
For more information about 10053, see Wolfgang Breitling’s http://www.hotsos.com/dnloads/1/10053/Breitling2002.pdf. Wolfgang will present this and another new 10053 paper at our Symposium in Dallas next February (http://www.hotsos.com/events/symposium).
Cary Millsap -----Original Message-----
I spent last two days working on that SQR and finally nailed it. Developers were using DISTINCT as a rule and CBO was choosing incorrect indexes.
I had to do following to make it work ...
1. we put hints in appropriate places 2. created 3 new indexes to help the queries 3. Created 1 MV 4. Change the pl/sql code to be more efficient 5. Remove unnecessary distinct conditions.
After all this said and done, I finally have the SQR taking around 375 seconds compared to about 1500 seconds. Even if I run all 8 of them in parallel it doesn't blow the TEMP segments.
Now comes the difficult part, to explain the development the things that we had to do and why we had to do them.
I think I am really starting to see the power of event 10046 ... I just wish I should learn more on event 10053 as well, but that will come too eventually.
I am still going to track the temp segment usage though ...
Thanks everyone for all your kind suggestions, I am not done yet, there is more work to be done.
Raj ______________________________________________________ Rajendra Jamadagni MIS, ESPN Inc. Rajendra dot Jamadagni at ESPN dot com Any opinion expressed here is personal and doesn't reflect that of ESPN Inc. QOTD: Any clod can have facts, but having an opinion is an art!
|
Title: Tracking Temp Segment Usage and Event 10046
- Tracking Temp Segment Usage and Event 10046 Jamadagni, Rajendra
- RE: Tracking Temp Segment Usage and Event 10046 K Gopalakrishnan
- RE: Tracking Temp Segment Usage and Event 10046 Jamadagni, Rajendra
- RE: Tracking Temp Segment Usage and Event 10046 DENNIS WILLIAMS
- RE: Tracking Temp Segment Usage and Event 10046 Khedr, Waleed
- RE: Tracking Temp Segment Usage and Event 10046 STEVE OLLIG
- RE: Tracking Temp Segment Usage and Event 10046 Jamadagni, Rajendra
- RE: Tracking Temp Segment Usage and Event 10046 Connor McDonald
- RE: Tracking Temp Segment Usage and Event 10046 Cary Millsap
- RE: Tracking Temp Segment Usage and Event 10046 Jamadagni, Rajendra
