To: Marc Cousin <cousinmarc@gmail.com>
cc: Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>,
	pgsql-performance@postgresql.org
Subject: Re: inheritance: planning time vs children number vs column number 
In-reply-to: <201102281409.55827.cousinmarc@gmail.com> 
References: <201102281038.04945.cousinmarc@gmail.com>
	<4D6B9BC9.1000404@enterprisedb.com>
	<201102281409.55827.cousinmarc@gmail.com>
Comments: In-reply-to Marc Cousin <cousinmarc@gmail.com>
	message dated "Mon, 28 Feb 2011 14:09:55 +0100"
Date: Mon, 28 Feb 2011 10:35:37 -0500
Message-ID: <7266.1298907337@sss.pgh.pa.us>
From: Tom Lane <tgl@sss.pgh.pa.us>

Marc Cousin <cousinmarc@gmail.com> writes:
> The Monday 28 February 2011 13:57:45, Heikki Linnakangas wrote :
>> Testing here with a table with 1000 columns and 100 partitions, about
>> 80% of the planning time is looking up the statistics on attribute
>> width, to calculate average tuple width. I don't see O(n^2) behavior,
>> though, it seems linear.

> It is only based on experimentation, for my part, of course… 

> If you measure the planning time, modifying either the columns or the 
> partitions number, the square root of the planning time is almost perfectly 
> proportional with the parameter you're playing with.

Could we see a concrete example demonstrating that?  I agree with Heikki
that it's not obvious what you are testing that would have such behavior.
I can think of places that would have O(N^2) behavior in the length of
the targetlist, but it seems unlikely that they'd come to dominate
runtime at a mere 1000 columns.

			regards, tom lane