Re: Query plan looks OK, but slow I/O - settings advice?

From: "Roger Hand" <RHand(at)kailea(dot)com>
To: "Jeffrey W(dot) Baker" <jwbaker(at)acm(dot)org>
Cc: <pgsql-performance(at)postgresql(dot)org>, "Leszek Kotzian" <lkotzian(at)ragingnet(dot)com>
Subject: Re: Query plan looks OK, but slow I/O - settings advice?
Date: 2005-08-19 07:35:23
Message-ID: DB28E9B548192448A4E8C8A3C1B1E475611C78@sj1-exch-01.us.corp.kailea.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Jeffrey W. Baker wrote:
> On Tue, 2005-08-16 at 10:46 -0700, Roger Hand wrote:
>> The disks are ext3 with journalling type of ordered, but this was later changed to writeback with no apparent change in speed.
>>
>> They're on a Dell poweredge 6650 with LSI raid card, setup as follows:
>> 4 disks raid 10 for indexes (145GB) - sdc1
>> 6 disks raid 10 for data (220GB) - sdd1
>> 2 mirrored disks for logs - sdb1
>>
>> stripe size is 32k
>> cache policy: cached io (am told the controller has bbu)
>> write policy: write-back
>> read policy: readahead
>
> I assume you are using Linux 2.6.

Oops, sorry I left that out. Nope, we're on 2.4:

[root(at)rage-db2 ~]$ uname -a
Linux xxx.xxx.xxx 2.4.21-27.0.2.ELsmp #1 SMP Wed Jan 12 23:35:44 EST 2005 i686 i686 i386 GNU/Linux

It's RedHat Enterprise AS3.0 Fri Nov 5 17:55:14 PST 2004

> Have you considered booting your
> machine with elevator=deadline?

I just did a little Googling and see that the 2.4 kernel didn't have a decent elevator tuning system, and that was fixed in 2.6. Hmmm ....

Thanks for the ideas ...

-Roger

Browse pgsql-performance by date

  From Date Subject
Next Message Mark Cotner 2005-08-19 08:24:04 sustained update load of 1-2k/sec
Previous Message Gavin Sherry 2005-08-19 07:17:45 Re: Query plan looks OK, but slow I/O - settings advice?