Re: Volunteer to build a configuration tool

From: Jim Nasby <decibel(at)decibel(dot)org>
To: Greg Smith <gsmith(at)gregsmith(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Volunteer to build a configuration tool
Date: 2007-06-26 00:19:01
Message-ID: BA5FEC30-926B-4FC5-BF17-E1BECE1B6BDD@decibel.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-docs pgsql-performance

On Jun 23, 2007, at 2:28 PM, Greg Smith wrote:
> On Thu, 21 Jun 2007, Campbell, Lance wrote:
>> I have a PostgreSQL database that runs on a dedicated server. The
>> server has 24Gig of memory. What would be the max size I would ever
>> want to set the shared_buffers to if I where to relying on the OS for
>> disk caching approach? It seems that no matter how big your
>> dedicated
>> server is there would be a top limit to the size of shared_buffers.
>
> It's impossible to say exactly what would work optimally in this
> sort of situation. The normal range is 25-50% of total memory, but
> there's no hard reason for that balance; for all we know your apps
> might work best with 20GB in shared_buffers and only a relatively
> small 4GB left over for the rest of the OS to use. Push it way up
> and and see what you get.
>
> This is part of why the idea of an "advanced" mode for this tool is
> suspect. Advanced tuning usually requires benchmarking with as
> close to real application data as you can get in order to make good
> forward progress.

Agreed. EnterpriseDB comes with a feature called "DynaTune" that
looks at things like server memory and sets a best-guess at a bunch
of parameters. Truth is, it works fine for 90% of cases, because
there's just a lot of installations where tuning postgresql.conf
isn't that critical.

The real issue is that the "stock" postgresql.conf is just horrible.
It was originally tuned for something like a 486, but even the recent
changes have only brought it up to the "pentium era" (case in point:
24MB of shared buffers equates to a machine with 128MB of memory,
give or take). Given that, I think an 80% solution would be to just
post small/medium/large postgresql.conf files somewhere.

I also agree 100% with Tom that the cost estimators need serious
work. One simple example: nothing in the planner looks at what
percent of a relation is actually in shared_buffers. If it did that,
it would probably be reasonable to extrapolate that percentage into
how much is sitting in kernel cache, which would likely be miles
ahead of what's currently done.
--
Jim Nasby jim(at)nasby(dot)net
EnterpriseDB http://enterprisedb.com 512.569.9461 (cell)

In response to

Responses

Browse pgsql-docs by date

  From Date Subject
Next Message Gregory Stark 2007-06-26 08:01:59 Re: Volunteer to build a configuration tool
Previous Message Greg Smith 2007-06-23 19:28:17 Re: Volunteer to build a configuration tool

Browse pgsql-performance by date

  From Date Subject
Next Message Stephen Frost 2007-06-26 00:33:39 Re:
Previous Message Ed Tyrrill 2007-06-26 00:09:58 Re: