Received: from localhost (mx1.hub.org [200.46.208.251]) by postgresql.org (Postfix) with ESMTP id 45E009FA4C5 for ; Thu, 12 Oct 2006 10:53:09 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.208.251]) (amavisd-new, port 10024) with ESMTP id 40055-10 for ; Thu, 12 Oct 2006 10:52:53 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from noel.decibel.org (noel.decibel.org [67.100.216.10]) by postgresql.org (Postfix) with ESMTP id 15A949FB39E for ; Thu, 12 Oct 2006 10:52:51 -0300 (ADT) Received: by noel.decibel.org (Postfix, from userid 1001) id 6C82F56431; Thu, 12 Oct 2006 08:52:50 -0500 (CDT) Received: (hashcash-sendmail, from uid 1001); Thu, 12 Oct 2006 08:52:49 -0500 Date: Thu, 12 Oct 2006 08:52:49 -0500 From: "Jim C. Nasby" To: Mark Kirkwood Cc: "H.J. Sanders" , pgsql-performance@postgresql.org Subject: Re: FW: Simple join optimized badly? Message-ID: <20061012135248.GS28647@nasby.net> References: <452E11FB.7070704@paradise.net.nz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <452E11FB.7070704@paradise.net.nz> X-Operating-System: FreeBSD 6.0-RELEASE-p4 amd64 X-Distributed: Join the Effort! http://www.distributed.net User-Agent: Mutt/1.5.11 X-Hashcash: 1:20:061012:markir@paradise.net.nz::hRMaDC8ArTqHkDEa:000000000000000 0000000000000000000000003hBa X-Hashcash: 1:20:061012:hjs@rmax.nl::P4g+yxErRFx1I44y:0000005IoF X-Hashcash: 1:20:061012:pgsql-performance@postgresql.org::cW1skOwccioadjcm:00000 0000000000000000000000008E8g X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0.05 tagged_above=0 required=5 tests=AWL X-Spam-Level: X-Archive-Number: 200610/231 X-Sequence-Number: 21360 On Thu, Oct 12, 2006 at 10:59:23PM +1300, Mark Kirkwood wrote: > H.J. Sanders wrote: > > > why not just like in some other (commercial) databases: > > > > a statement to say: use index ............ > > > > I know this is against all though but if even the big ones can not resist > > the pressure of their users, why not? > > > > Yeah - some could not (e.g. Oracle), but some did (e.g. DB2), and it > seemed (to me anyway) significant DB2's optimizer worked much better > than Oracle's last time I used both of them (Oracle 8/9 and DB2 7/8). If someone's going to commit to putting effort into improving the planner then that's wonderful. But I can't recall any significant planner improvements since min/max (which I'd argue was more of a bug fix than an improvement). In fact, IIRC it took at least 2 major versions to get min/max fixed, and that was a case where it was very clear-cut what had to be done. -- Jim Nasby jim@nasby.net EnterpriseDB http://enterprisedb.com 512.569.9461 (cell)