|
[oracle-l]
||
[Date Prev]
[12-2005 Date Index]
[Date Next]
||
[Thread Prev]
[12-2005 Thread Index]
[Thread Next]
Re: CBO - hash join vs nested loops
- From: "Radoulov, Dimitre" <cichomitiko@xxxxxxxxx>
- To: "Christian Antognini" <Christian.Antognini@xxxxxxxxxxxx>
- Date: Fri, 2 Dec 2005 17:44:42 +0100
Checked the CDN and NDV for the tables/columns and they seem OK, but I have
to check the indexes stats too.
The only think I noticed is that the CBO expects 2076 rows before the final
distinct and they are 1401.
I thought it could be something more "obvious" and "easy" :O).
Thanks,
Dimitre Radoulov
----- Original Message -----
From: "Christian Antognini" <Christian.Antognini@xxxxxxxxxxxx>
To: <cichomitiko@xxxxxxxxx>
Cc: <oracle-l@xxxxxxxxxxxxx>
Sent: Friday, December 02, 2005 4:32 PM
Subject: RE: CBO - hash join vs nested loops
Dimitre
Subject: CBO - hash join vs nested loops
Did you compare the estimated cardinalities with the real one? If not, it's
the right place to start... In fact with 3 conditions it's quite easy to
mislead the CBO.
HTH
Chris
--
http://www.freelists.org/webpage/oracle-l
|
|
