From: | "Thomas T(dot) Thai" <tom(at)minnesota(dot)com> |
---|---|
To: | Doug McNaught <doug(at)wireboard(dot)com> |
Cc: | PostgreSQL General <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: help with getting index scan |
Date: | 2002-02-25 16:23:05 |
Message-ID: | Pine.NEB.4.43.0202251021520.26912-100000@ns01.minnesota.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On 25 Feb 2002, Doug McNaught wrote:
> "Thomas T. Thai" <tom(at)minnesota(dot)com> writes:
>
> > On 25 Feb 2002, Doug McNaught wrote:
> > > Well, EXPLAIN is indicating (unless I misread it) that the estimate of
> > > rows returned is 336702, so it's not surprising that it opts for a
> > > sequential scan. Is this under 7.1 or 7.2? The latter keeps much
> > > better statistics about table populations...
> >
> > this is under 7.2. is there away to force it to use index scan? cause
> > right now when i'm searching using a cat reference, it's taking a few
> > seconds.
>
> I'm still suspicious that something is wrong, but you can do
>
> SET enable_seqscan TO off;
>
> before your query and see if it helps your performance. If it makes a
> significant difference let us know--Tom may be interested in trying to
> improve the statistics.
i just tried turning off seq scan and the query still takes up to 8
seconds which is 7 seconds too long. btw, how do you clean the cache
from the last query?
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2002-02-25 16:30:41 | Re: help with getting index scan |
Previous Message | Stephan Szabo | 2002-02-25 16:20:30 | Re: Work Around For Oracle Feature |