pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Uwe Bartels <uwe.bartels@gmail.com>
To: pgsql-performance@postgresql.org
Subject: how to find the sql using most of the i/o in an oltp system
Date: Sun, 15 Jun 2008 15:48:32 +0200
Message-ID: <48551DB0.8050507@gmail.com> (raw)

Hi,

I read a lot about postgres tuning and did some of it. But one of the 
things, when you start tuning a system that is completely new to you, is 
checking which sql statement(s) cost most of the resources.

cpu instensive sql seems easy to find.
But how do I find i/o intensive sql as fast as possible?

Tuning a sql statements I'm familiar with. Finding a sql statement which 
takes too long due to i/o is probably easy as well. But how about 
statements that take about 100 ms, that read a lot and that are executed 
several times per second?

To ask the same question just different - Is there a possibility to 
check how many pages/kB a sql reads from shared buffer and how many 
pages/kB from disk?
Is there a possibility to read or create historical records about how 
often, how fast one sql is run and how many reads it used from shared 
buffers, from disk an so on.


Best regards,
Uwe



view thread (3+ messages)  latest in thread

Message-ID: <48551DB0.8050507@gmail.com>
Permalink:  ../48551DB0.8050507@gmail.com/
Also on:    postgresql.org/message-id/48551DB0.8050507@gmail.com

 · 

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: pgsql-performance@postgresql.org
  Cc: uwe.bartels@gmail.com
  Subject: Re: how to find the sql using most of the i/o in an oltp system
  In-Reply-To: <48551DB0.8050507@gmail.com>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox