Re: performance config help

From: Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com>
To: Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>
Cc: Bob Dusek <redusek(at)gmail(dot)com>, pgsql-performance(at)postgresql(dot)org
Subject: Re: performance config help
Date: 2010-01-11 19:38:42
Message-ID: dcc563d11001111138t5a2003b8j736e9f5beacf7636@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Mon, Jan 11, 2010 at 12:36 PM, Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com> wrote:
> FYI, on an 8 or 16 core machine, 10k to 30k context switches per
> second aren't that much.  If you're climbing past 100k you might want
> to look out.
>
> The more I read up on the 74xx CPUs and look at the numbers here the
> more I think it's just that this machine has X bandwidth and it's
> using it all up.  You could put 1,000 cores in it, and it wouldn't go
> any faster.  My guess is that a 4x6 core AMD machine or even a 2x6
> Nehalem would be much faster at this job.  Only way to tell is to run
> something like the stream benchmark and see how it scales,
> memory-wise, as you add cores to the benchmark.

Also I'm guessing that query profiling may help, if we can get the
queries to request less data to trundle through then we might be able
to get Bob the performance needed to keep up.

But at some point he's gonna have to look at partitioning his database
onto multiple machines some how.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Jeremy Harris 2010-01-11 19:41:08 Re: Choice of bitmap scan over index scan
Previous Message Scott Marlowe 2010-01-11 19:36:39 Re: performance config help