Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> RE: When stats trash your performance

RE: When stats trash your performance

From: Steve Ollig <sollig_at_lifetouch.com>
Date: Wed, 7 Dec 2005 13:09:28 -0600
Message-ID: <DFA54221DF26D911BC5100B0D079D97A072ABEEE@EXCHMN5>


Stephen -  

I believe the reasoning behind that school of thought is that statistics should adhere to normal change management procedures just as application code would. You wouldn't sling a code change into production without running it through a well defined QA process. Why wouldn't you do the same with statistics?

-----Original Message-----
From: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org]On Behalf Of stephen booth
Sent: Wednesday, December 07, 2005 12:59 PM To: oracle-l_at_freelists.org
Subject: When stats trash your performance

I've heard mention in a number of talks that you shouldn't gather stats very often, even not at all once a system is bedded in, as it can cause your database to start performing really badly. I don't recall any one ever saying, or ever reading, an explanation as to why this might happen. Perhaps I'm looking at it from the wrong perspective but it seems axiomic that the fresher your stats the better decisions CBO is likely to make.



This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email
--
http://www.freelists.org/webpage/oracle-l
Received on Wed Dec 07 2005 - 13:24:01 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US