Re: Performance problems on 4/8way Opteron (dualcore) HP

From: Dirk Lutzebäck <lutzeb(at)aeccom(dot)com>
To: "J(dot) Andrew Rogers" <jrogers(at)neopolitan(dot)com>
Cc: josh(at)agliodbs(dot)com, pgsql-performance(at)postgresql(dot)org, Sven Geisler <sgeisler(at)aeccom(dot)com>
Subject: Re: Performance problems on 4/8way Opteron (dualcore) HP
Date: 2005-07-31 10:12:32
Message-ID: 42ECA410.7020804@aeccom.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Anybody knows if RedHat is already supporting this patch on an
enterprise version?

Regards,

Dirk

J. Andrew Rogers wrote:
> On 7/29/05 10:46 AM, "Josh Berkus" <josh(at)agliodbs(dot)com> wrote:
>
>>>does anybody have expierence with this machine (4x 875 dual core Opteron
>>>CPUs)?
>>
>>Nope. I suspect that you may be the first person to report in on
>>dual-cores. There may be special compile issues with dual-cores that
>>we've not yet encountered.
>
>
>
> There was recently a discussion of similar types of problems on a couple of
> the supercomputing lists, regarding surprisingly substandard performance
> from large dual-core opteron installations.
>
> The problem as I remember it boiled down to the Linux kernel handling
> memory/process management very badly on large dual core systems --
> pathological NUMA behavior. However, this problem has apparently been fixed
> in Linux v2.6.12+, and using the more recent kernel on large dual core
> systems generated *massive* performance improvements on these systems for
> the individuals with this issue. Using the patched kernel, one gets the
> performance most people were expecting.
>
> The v2.6.12+ kernels are a bit new, but they contain a very important
> performance patch for systems like the one above. It would definitely be
> worth testing if possible.
>
>
> J. Andrew Rogers
>
>

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message J. Andrew Rogers 2005-07-31 15:29:15 Re: Performance problems on 4/8way Opteron (dualcore)
Previous Message Dirk Lutzebäck 2005-07-31 10:11:02 Re: Performance problems on 4/8way Opteron (dualcore) HP