Re: Urgent Request

From: Mark D Powell <Mark.Powell_at_eds.com>
Date: Tue, 26 Feb 2008 09:13:25 -0800 (PST)
Message-ID: <0bc92fdd-98b0-49bf-91b1-c8148565680c@60g2000hsy.googlegroups.com>


On Feb 26, 8:59 am, spamb..._at_milmac.com (Doug Miller) wrote:
> In article <e3cf774b-d243-49e4-9436-48cca7388..._at_e6g2000prf.googlegroups.com>, Smitty <marksmith..._at_hotmail.com> wrote:
>
> >Hello everyone.  I am really hoping that someone can help me with this
> >issue.  I have two identical databases (one is a replication of the
> >other), where when I run an identical query on each, one returns the
> >results in .5 seconds, while the other takes 18 seconds to return the
> >same result.  The indexes etc. are all identical on each database.
> >When we ran a trace, we noticed that the disk gets hit only once on
> >the 'quick' database, but get hit over 10000 times on the 'slow'
> >database to return the same number of rows.  Does anyone have any idea
> >what could be the issue.  I am attaching the query below.  Thank you
> >very much.
>
> Statistics haven't been updated recently on the 'slow' database?

Besides the age or sample size of the statistics with version 9+ and depending on if any of the constants shown in the OP are bind variables or not, the bind variable peeking feature could also be an issue.

Run an explain plan on both servers and compare them.

HTH -- Mark D Powell -- Received on Tue Feb 26 2008 - 11:13:25 CST

Original text of this message