From: Mark Kirkwood <markir@paradise.net.nz>
To: Dan Gorman <dgorman@hi5.com>
Cc: Luke Lonergan <llonergan@greenplum.com>
Cc: Joshua D. Drake <jd@commandprompt.com>
Cc: Craig A. James <cjames@modgraph-usa.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Reliability recommendations
Date: Sat, 25 Feb 2006 14:18:31 +1300
Message-ID: <43FFB067.2070406@paradise.net.nz> (raw)
In-Reply-To: <125CA0A8-6142-4DF5-822E-CF0B0C46B4C1@hi5.com>
References: <C024DFA1.1DC2D%llonergan@greenplum.com>
<43FFA503.5070700@paradise.net.nz>
<125CA0A8-6142-4DF5-822E-CF0B0C46B4C1@hi5.com>
Dan Gorman wrote:
> All,
>
> Was that sequential reads? If so, yeah you'll get 110MB/s? How big was
> the datafile size? 8MB? Yeah, you'll get 110MB/s. 2GB? No, they can't
> sustain that. There are so many details missing from this test that
> it's hard to have any context around it :)
>
Actually they can. Datafile size was 8G, machine had 2G RAM (i.e.
datafile 4 times memory). The test was for a sequential read with 8K
blocks. I believe this is precisely the type of test that the previous
posters were referring to - while clearly, its not a real-world measure,
we are comparing like to like, and as such terrible results on such a
simple test are indicative of something 'not right'.
regards
Mark
P.s. FWIW - I'm quoting a test from a few years ago - the (same) machine
now has 4 RAID0 ata disks and does 175MB/s on the same test....
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: markir@paradise.net.nz, dgorman@hi5.com, llonergan@greenplum.com, jd@commandprompt.com, cjames@modgraph-usa.com
Subject: Re: Reliability recommendations
In-Reply-To: <43FFB067.2070406@paradise.net.nz>
* 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