Skip site navigation (1) Skip section navigation (2)

Re: Disk and shared buffer cache use

From: "Chamara Gunaratne" <chamarag(at)gmail(dot)com>
To: pgsql-novice(at)postgresql(dot)org
Subject: Re: Disk and shared buffer cache use
Date: 2007-05-10 17:54:15
Message-ID: db0a7ed90705101054xd548693g44cb26b2952ac51f@mail.gmail.com (view raw or flat)
Thread:
Lists: pgsql-novice
Thanks guys, you comments are much appreciated. My background is
mainly in network performance evaluation and I am new to db tuning. I
will enable block level stats and get the data.

The reason I am looking for this data is that I am debugging a very
slow db server. CPU wait time % is over 50% and it is constantly
reading in MB's of data (vmstat bi column). I have shared buffers at
128MB  and effective cache size at 256MB. and the db size is around
1.5GB. I am guessing that the db working set is much larger than
available cache memory.

Cheers,

-- Chamara

On 5/9/07, Jim Nasby <decibel(at)decibel(dot)org> wrote:
> On May 8, 2007, at 2:29 PM, Chamara Gunaratne wrote:
> > Hello All,
> > Is there a way to determine the percentage of data found in the shared
> > buffer cache when running a query in PostgreSQL 7.4? Or to put it in
> > another way, how do you measure the effectiveness of the shared buffer
> > cache?
> >
> > I read through several sites on PostgreSQL performance tuning, but
> > couldn't find a method to measure this.
> > Thank you,
>
> Look at the pg_statio_* views (you'll need block level stats enabled
> in postgresql.conf).
>
> Personally, I wouldn't worry too much about this unless you're trying
> to chase some specific performance issue.
> --
> Jim Nasby                                            jim(at)nasby(dot)net
> EnterpriseDB      http://enterprisedb.com      512.569.9461 (cell)
>
>
>


-- 
Priyanga Chamara Gunaratne

In response to

pgsql-novice by date

Next:From: chester c youngDate: 2007-05-10 18:34:22
Subject: Re: Count rows by day interval
Previous:From: Richard Broersma JrDate: 2007-05-10 15:33:55
Subject: Re: Count rows group by time intervals

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group