Re: Performance weirdness with/without vacuum analyze

From: Greg Stark <gsstark(at)mit(dot)edu>
To: pgsql-general(at)postgresql(dot)org
Subject: Re: Performance weirdness with/without vacuum analyze
Date: 2003-10-13 17:40:11
Message-ID: 87fzhx2g04.fsf@stark.dyndns.tv
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general


Harry Broomhall <harry(dot)broomhall(at)uk(dot)easynet(dot)net> writes:

> I'm not entirely sure how I would do this, as the 'lookup' is actualy
> a join. I thought that the order of nameing the joined tables didn't
> matter (except for 'left' and 'right'), similar to the fact that 1 + 2 is
> the same as 2 + 1.

Outer joins are commutative (1+2 == 2+1) at least barring the left/right
issue. But they are not associative. (1+2)+3 != 1+(2+3).

There are cases where the optimizer could prove a particular set of joins
could be reordered this way, but it doesn't know how to do that currently.

--
greg

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Patrick Welche 2003-10-13 18:22:34 gborg cvs ?
Previous Message Jan Wieck 2003-10-13 16:06:58 Re: more on undefined reference to 'pg_detoast_datum'