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

Home -> Community -> Usenet -> c.d.o.server -> Re: Correlated Subquery Performance Puzzle...

Re: Correlated Subquery Performance Puzzle...

From: Spendius <spendius_at_muchomail.com>
Date: 20 Dec 2005 05:40:41 -0800
Message-ID: <1135086041.610322.50020@g44g2000cwa.googlegroups.com>


mccmx_at_hotmail.com schreef:
> Why does Oracle decide to use a hash join instead of a NL join when I
> add this new WHERE clause into the subquery block...?
Because the optimizer finds the hash is cheaper than the nl when you add "AND asg.effdt <= cal.dur". Check it with a 10053 trace, you'll see all steps followed by the optimizer.

I'm uncapable of telling you why it's so, but if you want to keep the nested
loop just hint your statement (but can you ?, as you're obviously dealing
with a PeopleSoft application...). Received on Tue Dec 20 2005 - 07:40:41 CST

Original text of this message

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