Can't remember if you got any responses on this - if you did, then I
apologize for any repetitions...
There's no correct answer to this question - if there was, it would be
part of the DBMS_STATS thing now, I bet :-)
I would suggest the following (subject to debate, please):
1. Collect stats (sample 1 percent) on all your objects OR
2. Import stats from your test system (or another similar system)
3. If there are performance problems somewhere, fix them with either SQL
statement tuning or Breitlings methods, or whatever
4. If a table grows a LOT in size, re-analyze.
5. Otherwise, don't touch anything until somebody complains. What Peter
Gram from Miracle has called Compulsive Analyze DisOrder (CADO) is
perhaps fun, but mostly not needed.
Why the 1 percent sample? Because that's what the Oracle benchmark guys
do, so I figured it might be good enough. From a purely statistical
point of view it ought to be, too.
Special problems exist, but this is my first suggestion.
Mogens
April Wells wrote:
Carel-Jan
It is a really good paper, one that I will now read over pretty extensively... but I'm not sure it answers her questions on DBMS_STATS and how best to collect statistics effectively.
I do want to see any good doc on it to. We are using DBMS_STATS in production, in all of our instances, but I'm not comfortable that we are implementing as well as we could, either.
ajw
April Wells
Oracle DBA/Oracle Apps DBA
Corporate Systems
Amarillo Texas
@>-->-->--
"Few people really enjoy the simple pleasure of flying a kite"
Adam Wells age 11
"Imagination is the highest kite one can fly."
Lauren Bacall
---------------------------------------------------------------- Please see the official ORACLE-L FAQ: http://www.orafaq.com ---------------------------------------------------------------- To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put 'unsubscribe' in the subject line. -- Archives are at //www.freelists.org/archives/oracle-l/ FAQ is at //www.freelists.org/help/fom-serve/cache/1.html -----------------------------------------------------------------
