Re: Very high effective_cache_size == worse performance? - Mailing list pgsql-performance

From Scott Marlowe
Subject Re: Very high effective_cache_size == worse performance?
Date
Msg-id w2gdcc563d11004201120h88b16774ve81b22f60a875c9c@mail.gmail.com
Whole thread Raw
In response to Re: Very high effective_cache_size == worse performance?  (Scott Marlowe <scott.marlowe@gmail.com>)
List pgsql-performance
On Tue, Apr 20, 2010 at 12:15 PM, Scott Marlowe <scott.marlowe@gmail.com> wrote:
> On Tue, Apr 20, 2010 at 11:39 AM, David Kerr <dmk@mr-paradox.net> wrote:
>> Howdy all,
>>
>> I've got a huge server running just postgres. It's got 48 cores and 256GB of ram. Redhat 5.4, Postgres 8.3.9.
>> 64bit OS. No users currently.
>
> What's your IO subsystem look like?  What did vmstat actually say?

Note that on a 48 core machine, if vmstat shows 2% wait and 98% idle
then you'd be 100% io bound, because it's % of total CPU.  iostat -x
10 will give a better view of how hard your disks are working, and if
they're the issue.

pgsql-performance by date:

Previous
From: Kris Jurka
Date:
Subject: Re: Very high effective_cache_size == worse performance?
Next
From: David Kerr
Date:
Subject: Re: Very high effective_cache_size == worse performance?