From: Chris Browne <cbbrowne@acm.org>
Subject: Re: Which processor runs better for Postgresql?
Date: Tue, 13 Jun 2006 14:22:06 -0400
Organization: cbbrowne Computing Inc
Lines: 36
Message-ID: <60bqswvru9.fsf@dba2.int.libertyrms.com>
References: <1150221760.12191.25.camel@amd64-gentoo-laptop>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
User-Agent: Gnus/5.1007 (Gnus v5.10.7) XEmacs/21.4.19 (linux)
Cancel-Lock: sha1:FQV0QgZOeRV/c/CzGiljFJc4sUc=
To: pgsql-performance@postgresql.org

steve.poe@gmail.com (Steve Poe) writes:
> I have a client who is running Postgresql 7.4.x series database
> (required to use 7.4.x). They are planning an upgrade to a new server.
> They are insistent on Dell.

Then they're being insistent on poor performance.

If you search for "dell postgresql performance" you'll find plenty of
examples of people who have been disappointed when they insisted on
Dell for PostgreSQL.

Here is a *long* thread on the matter...
<http://archives.postgresql.org/pgsql-performance/2004-12/msg00022.php>

> I am hoping the client is willing to wait for Dell to ship a AMD
> Opeteron-based server.

Based on Dell's history, I would neither:

 a) Hold my breath, nor

 b) Expect an Opteron-based Dell server to perform as well as
    seemingly-equivalent servers provisioned by other vendors.

We got burned by some Celestica-built Opterons that didn't turn out
quite as hoped.

We have had somewhat better results with some HP Opterons; they appear
to be surviving less-than-ideal 3rd world data centre situations with
reasonable aplomb.  (Based on the amount of dust in their diet, I'm
somewhat surprised the disk drives are still running...)
-- 
let name="cbbrowne" and tld="acm.org" in name ^ "@" ^ tld;;
http://cbbrowne.com/info/nonrdbms.html
We are Pentium of Borg.  Division is futile. You will be approximated.
(seen in someone's .signature)