From: Jim C. Nasby <jnasby@pervasive.com>
To: Joost Kraaijeveld <J.Kraaijeveld@Askesis.nl>
Cc: Pgsql-Performance (E-mail) <pgsql-performance@postgresql.org>
Subject: Re: Can anyone explain this pgbench results?
Date: Tue, 7 Mar 2006 13:09:34 -0600
Message-ID: <20060307190933.GY82989@pervasive.com> (raw)
In-Reply-To: <A3D1526C98B7C1409A687E0943EAC410605FC1@obelix.askesis.nl>
References: <A3D1526C98B7C1409A687E0943EAC410605FC1@obelix.askesis.nl>
On Mon, Mar 06, 2006 at 04:29:49PM +0100, Joost Kraaijeveld wrote:
> I was expecting a low(ish) score the first run (because the tables are not loaded in the cache yet), followed by continues high(ish) scores, but not an alternating pattern. I also did not expect so much difference, given the hardware I have (Dual Opteron, 4GB memory , 3Ware SATA RAID5 with 5 disks, seerate swap and pg_log disks).
On a side-note:
RAID5 and databases generally don't mix well.
Most people find that pg_xlog will live happily with the OS; it's the
data files that need the most bandwidth.
If you start swapping, performance will tank to the point that it's
unlikely that swap being on seperate disks will help at all. Better off
to just keep it with the OS and use the disks for the database tables.
Speaking of 'disks', what's your exact layout? Do you have a 5 drive
raid5 for the OS and the database, 1 drive for swap and 1 drive for
pg_xlog?
--
Jim C. Nasby, Sr. Engineering Consultant jnasby@pervasive.com
Pervasive Software http://pervasive.com work: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: jnasby@pervasive.com, J.Kraaijeveld@Askesis.nl
Subject: Re: Can anyone explain this pgbench results?
In-Reply-To: <20060307190933.GY82989@pervasive.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox