From: | "Ross J(dot) Reedstrom" <reedstrm(at)rice(dot)edu> |
---|---|
To: | Tim Perdue <tperdue(at)valinux(dot)com> |
Cc: | pgsql-hackers(at)hub(dot)org |
Subject: | Re: 7.0.2 issues / Geocrawler |
Date: | 2000-07-12 15:00:08 |
Message-ID: | 20000712100008.A1487@rice.edu |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Wed, Jul 12, 2000 at 06:17:23AM -0700, Tim Perdue wrote:
> Mike Mascari wrote:
> > Have you VACUUM ANALYZE'd the table(s) in question?
>
> Yes, they've been vacuum analyze'd and re-vaccum analyze'd to death.
> Also added some extra indexes that I don't really need just to see if
> that helps.
Tim, why are you building a multikey index, especially one containing a
large text field? It's almost never a win to index a text field, unless
all the WHERE clauses that use it are either anchored to the beginning
of the field, or are equality tests (in which case, the field is really
an enumerated type, masquerading as a text field)
A multikey index is only useful for a very limited set of queries. Here's
a message from last August, where Tom Lane talks about that:
http://www.postgresql.org/mhonarc/pgsql-sql/1999-08/msg00145.html
Ross
--
Ross J. Reedstrom, Ph.D., <reedstrm(at)rice(dot)edu>
NSBRI Research Scientist/Programmer
Computer and Information Technology Institute
Rice University, 6100 S. Main St., Houston, TX 77005
From | Date | Subject | |
---|---|---|---|
Next Message | Philip Warner | 2000-07-12 15:11:04 | RE: RE: [HACKERS] pg_dump & blobs - editable dump? |
Previous Message | The Hermit Hacker | 2000-07-12 14:57:19 | Re: Vacuum only with 20% old tuples |