To: Andrew Sullivan <ajs@crankycanuck.ca>
cc: pgsql-performance@postgresql.org
Subject: Re: TIMING A QUERY ??? 
In-reply-to: <20070711153911.GQ1241@phlogiston.dyndns.org> 
References: <11542393.post@talk.nabble.com>
	<20070711153911.GQ1241@phlogiston.dyndns.org>
Comments: In-reply-to Andrew Sullivan <ajs@crankycanuck.ca>
	message dated "Wed, 11 Jul 2007 11:39:11 -0400"
Date: Wed, 11 Jul 2007 12:10:55 -0400
Message-ID: <26707.1184170255@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Andrew Sullivan <ajs@crankycanuck.ca> writes:
> On Wed, Jul 11, 2007 at 08:21:40AM -0700, smiley2211 wrote:
>> How can I get the time it takes a query to execute - explain analyze is
>> taking over 5 hours to complete

> You can't get it any faster than what explain analyse does: it runs
> the query.  How else would you get the answer?

Well, on some platforms (ie consumer-grade PCs) explain analyze can be a
lot slower than just running the query, because of the overhead of all
those gettimeofday() calls it does.  El cheapo clock hardware is slow
to read.  (I think the problem is actually that the PC-standard hardware
API for clocks was designed back when taking a whole microsecond to read
the clock didn't seem like a problem.)

>> ...can I use \timing???  I don't get any time when using the
>> \timing option...

> How so?  It returns Time: N ms at the end of output for me.

Works for me too.

			regards, tom lane