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

Re: Reason for PG being seemingly I/O bound?

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Harshad <harshad(dot)rj(at)gmail(dot)com>
Cc: pgsql-novice(at)postgresql(dot)org
Subject: Re: Reason for PG being seemingly I/O bound?
Date: 2009-09-13 19:15:27
Message-ID: 8970.1252869327@sss.pgh.pa.us (view raw or flat)
Thread:
Lists: pgsql-novice
Harshad <harshad(dot)rj(at)gmail(dot)com> writes:
> I am trying to figure out why a very simple query turns out to be I/O
> bound.

It's hard to see how that case could be I/O bound, unless your machine
is seriously starved for memory.  Did you watch "vmstat 1" or something
similar to confirm that a lot of I/O is really happening?

> I have put this query in simple.sql, and running it in a tight loop in the shell thusly:

> while [ true ] ; do psql -d twinkle -U postgres -f simple.sql; done > /dev/null

FWIW, the overhead of the above is just enormous --- starting up a new
database connection takes way more time than issuing one simple query.
However I still don't see how it would be I/O bound; the kernel
certainly ought to have everything needed in disk cache after a couple
of cycles.  On my machine a similar test immediately pins the CPU
with about half user, half system time.

What platform are you using, exactly?

			regards, tom lane

In response to

Responses

pgsql-novice by date

Next:From: Sean DavisDate: 2009-09-14 01:42:47
Subject: Re: using SQL for multi-machine job management?
Previous:From: manausDate: 2009-09-13 17:38:19
Subject: qualified names

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