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: Index Compression question - afterquestion

Re: Index Compression question - afterquestion

From: Niall Litchfield <n-litchfield_at_audit-commission.gov.uk>
Date: Tue, 21 Jan 2003 11:33:34 -0000
Message-ID: <3e2d300f$0$246$ed9e5944@reading.news.pipex.net>


"Andy" <andy.spaven_at_eps-hq.co.uk> wrote in message news:GJ9X9.3579$9R.12250166_at_newsr2.u-net.net...
> Howard / Niall / Jan
>
> Yep - just checked 8.1.7 as well - it doesn't appear to wipe stats on a
> rebuild. There goes another assumption lurking as a false training memory
> :-). Just goes to show. That now leaves Jan's question about poor
> performance after the rebuild unanswered. If it was okay before and we
know
> the stats haven't changed what caused a slow down ?

Actually we know that the stats HAVE changed (if i read jan's post correctly.) He states that he ran

alter index rebuild compress;
followed by analyze compute.

I wouldn't expect typical explains to change but it is always possible, I also wonder if anything else significant has changed since the last analyze *prior* to the rebuild - like large volumes of data being added/deleted. I certainly can't see that running DBMS_STATS instead of ANALYZE would make any difference to the stats actually gathered.

--
Niall Litchfield
Oracle DBA
Audit Commission UK
*****************************************
Please include version and platform
and SQL where applicable
It makes life easier and increases the
likelihood of a good answer

******************************************
Received on Tue Jan 21 2003 - 05:33:34 CST

Original text of this message

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