pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  

help / color / mirror / Atom feed
From: Greg Smith <gsmith@gregsmith.com>
To: André Volpato <andre.volpato@ecomtecnologia.com.br>
Cc: pgsql-performance@postgresql.org
Subject: Re: Postgres not using array
Date: Thu, 28 Aug 2008 22:29:45 -0400 (EDT)
Message-ID: <Pine.GSO.4.64.0808282215220.11207@westnet.com> (raw)
In-Reply-To: <48AD735D.6040403@ecomtecnologia.com.br>
References: <48AC8292.5000708@ecomtecnologia.com.br>
	<48AD735D.6040403@ecomtecnologia.com.br>

On Thu, 21 Aug 2008, André Volpato wrote:

> So, what is slowing down is the CPU (Intel(R) Pentium(R) Dual CPU E2160 
> @ 1.80GHz)..In practice, I have noticed that dual 1.8 is worse than 
> single 3.0. We have another server wich is a Pentium D 3.0 GHz, that 
> runs faster.

Pentium D models are all dual-core so either you've got the wrong model 
number here or you've actually comparing against a 2X3.0GHz part.

The Core 2 Duo E2160 has a very small CPU cache--512KB per core.  Your 
older Pentium system probably has quite a bit more.  I suspect that's the 
main reason it runs faster on this application.

> I am a bit confused about what CPU is best for Postgres. Our apps is 
> mostly read, with a few connections and heavy queryes.

There are a lot of things you can run into with Postgres that end up being 
limited by the fact that they only run on a single core, as you've seen 
here.  If you've only got a fairly small number of connections running CPU 
heavy queries, you probably want a processor with lots of L2 cache and a 
fast clock speed, rather than adding a large number of cores running at a 
slower speed.  The very small L2 cache on your E2160 is likely what's 
holding it back here, and even though the newer processors are 
significantly more efficient per clock the gap between 1.8GHz and 3.0GHz 
is pretty big.

--
* Greg Smith gsmith@gregsmith.com http://www.gregsmith.com Baltimore, MD


view thread (12+ messages)

Message-ID: <Pine.GSO.4.64.0808282215220.11207@westnet.com>
Permalink:  ../Pine.GSO.4.64.0808282215220.11207@westnet.com/
Also on:    postgresql.org/message-id/Pine.GSO.4.64.0808282215220.11207@westnet.com

 · 

reply
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: gsmith@gregsmith.com, andre.volpato@ecomtecnologia.com.br
  Subject: Re: Postgres not using array
  In-Reply-To: <Pine.GSO.4.64.0808282215220.11207@westnet.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