From: | Lorenz Bateman <lorenz(at)ivision(dot)co(dot)uk> |
---|---|
To: | Norman Clarke <norman(at)combimatrix(dot)com> |
Cc: | <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Speed issues with update |
Date: | 2000-11-08 12:21:14 |
Message-ID: | Pine.GSO.4.30.0011081219001.25533-100000@avengers |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Yup, a full VACUUM analyze was done on the table. The problem seems to be
that 95% of processor time is spent in the kernal opening and closing the
same table file. This does seem a little strange. Could this be a problem
with the code? I've also tried a full reinstall, but to no avail.
Loz
On Thu, 2 Nov 2000, Norman Clarke wrote:
> Lorenz,
>
> Have you run VACUUM on the table?
>
> Norman
>
> On Wednesday 01 November 2000 05:43 am, you wrote:
> > I'm using 'PostgreSQL 7.0.2 on i386-unknown-freebsdelf4.0, compiled by
> > cc'.
> >
> > The problem is that the following statement takes approx 3hrs on a table
> > with approx 136,000 records.
> >
> > update table_name set state = 10 where pid is null;
> >
> > pid is indexed. The result from an explain on the statement is as
> > follows...
> >
> > Seq Scan on table_name (cost=0.00..8668.56 rows=45348 width=310)
> >
> > When the query is running, disk io is minimal, but cpu is maxed.
> >
> > Can anyone suggest a solution or a way of finding where pg is spending all
> > the cpu time?
> >
> > Thanks
> >
> > Loz
>
From | Date | Subject | |
---|---|---|---|
Next Message | Martin A. Marques | 2000-11-08 13:36:58 | TEXT and BLOBS |
Previous Message | Michaël Fiey | 2000-11-08 10:05:50 | Re: PGBUFFERS |