Re: Index creation and maintenance_work_mem

From: Alan Hodgson <ahodgson(at)simkin(dot)ca>
To: pgsql-general(at)postgresql(dot)org
Subject: Re: Index creation and maintenance_work_mem
Date: 2008-07-29 19:19:14
Message-ID: 200807291219.15083@hal.medialogik.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Tuesday 29 July 2008, "Francisco Reyes" <lists(at)stringsutils(dot)com> wrote:
> On 2:53 pm 07/29/08 Alan Hodgson <ahodgson(at)simkin(dot)ca> wrote:
> > --sar 2 30
>
> Linux 2.6.9-42.ELsmp (trans03) 07/29/2008
>
> 12:58:09 PM CPU %user %nice %system %iowait %idle
> 12:58:11 PM all 12.44 0.00 0.06 0.00 87.50
> 12:58:13 PM all 12.44 0.00 0.00 0.06 87.50
> 12:58:15 PM all 12.49 0.00 0.06 0.00 87.45
> 12:58:17 PM all 12.43 0.00 0.06 1.62 85.88
> 12:58:19 PM all 12.38 0.00 0.06 0.00 87.55
> 12:58:21 PM all 12.43 0.00 0.12 0.00 87.45
> 12:58:23 PM all 12.50 0.00 0.00 0.00 87.50
> 12:58:25 PM all 12.42 0.00 0.12 0.00 87.45
>
> Quadcore machine. Not using not even 100% of one core.
> Disk subsystem doesn't seem to be issue (no iowait).

That looks an awful lot like pegged 1/8th usage to me ... are you sure you
only have 4 cores? Hyper-threading?

>
> Based on a few links I read I thought that, holding everything else
> constant, increasing maintenance_work_mem would have helped.

Yeah, but it didn't.

--
Alan

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Francisco Reyes 2008-07-29 19:34:32 Re: Index creation and maintenance_work_mem
Previous Message Francisco Reyes 2008-07-29 19:03:35 Re: Index creation and maintenance_work_mem