Re: Adding more memory = hugh cpu load - Mailing list pgsql-performance

From Samuel Gendler
Subject Re: Adding more memory = hugh cpu load
Date
Msg-id CAEV0TzD8uae+1yqK3Z9XYL9_S57ZJHz3w-UirHaCvSj=XodiYQ@mail.gmail.com
Whole thread Raw
In response to Re: Adding more memory = hugh cpu load  ("Kevin Grittner" <Kevin.Grittner@wicourts.gov>)
Responses Re: Adding more memory = hugh cpu load
Re: Adding more memory = hugh cpu load
List pgsql-performance
On Mon, Oct 10, 2011 at 1:52 PM, Kevin Grittner <Kevin.Grittner@wicourts.gov> wrote:
alexandre - aldeia digital <adaldeia@gmail.com> wrote:

> I came to the list to see if anyone else has experienced the same
> problem

A high load average or low idle CPU isn't a problem, it's a
potentially useful bit of information in diagnosing a problem.  I
was hoping to hear what the actual problem was, since I've had a few
problems in high RAM situations, but the solutions depend on what
the actual problems are.  I don't suppose you saw periods where
queries which normally run very quickly (say in a millisecond or
less) were suddenly taking tens of seconds to run -- "stalling" and
then returning to normal?  Because if I knew you were having a
problem like *that* I might have been able to help.  Same for other
set of symptoms; it's just the suggestions would have been
different.  And the suggestions would have depended on what your
system looked like besides the RAM.

If you're satisfied with how things are running with less RAM,
though, there's no need.

The original question doesn't actually say that performance has gone down, only that cpu utilization has gone up. Presumably, with lots more RAM, it is blocking on I/O a lot less, so it isn't necessarily surprising that CPU utilization has gone up.  The only problem would be if db performance has gotten worse. Maybe I missed a message where that was covered?  I don't see it in the original query to the list.

pgsql-performance by date:

Previous
From: "Kevin Grittner"
Date:
Subject: Re: Adding more memory = hugh cpu load
Next
From: Claudio Freire
Date:
Subject: Re: Adding more memory = hugh cpu load