Re: Performance Tuning

From: Mike Rylander <mrylander(at)gmail(dot)com>
To: Chris Kratz <chris(dot)kratz(at)vistashare(dot)com>
Cc: Merlin Moncure <merlin(dot)moncure(at)rcsonline(dot)com>, pgsql-performance(at)postgresql(dot)org
Subject: Re: Performance Tuning
Date: 2005-02-10 01:27:16
Message-ID: b918cf3d0502091727267dd57d@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Wed, 9 Feb 2005 17:30:41 -0500, Chris Kratz
<chris(dot)kratz(at)vistashare(dot)com> wrote:
> The solutions appear to primarily be:
> 1. Going to faster hardware of which probably Opterons would be about the only
> choice. And even that probably won't be a huge difference.

I'd beg to differ on that last part. The difference between a 3.6GHz
Xeon and a 2.8GHz Opteron is ~150% speed increase on the Opteron on my
CPU bound app. This is because the memory bandwidth on the Opteron is
ENORMOUS compared to on the Xeon. Add to that the fact that you
actually get to use more than about 2G of RAM directly and you've got
the perfect platform for a high speed database on a budget.

> 2. Moving to more materialized views and prepared statements where we can.

Definitely worth investigating. I wish I could, but I can't get my
customers to even consider slightly out of date stats.... :(

> 3. Continue to tweak the sql behind our app.

Short of an Opteron based system, this is by far your best bet.

--
Mike Rylander
mrylander(at)gmail(dot)com
GPLS -- PINES Development
Database Developer
http://open-ils.org

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Iain 2005-02-10 01:50:16 Re: How can I make use of both CPUs in a dual processor
Previous Message Neil Conway 2005-02-10 01:13:03 Re: GiST indexes and concurrency (tsearch2)