X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (av.hub.org [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 445C9D7E67 for ; Wed, 9 Nov 2005 20:54:37 -0400 (AST) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 55962-10 for ; Thu, 10 Nov 2005 00:54:34 +0000 (GMT) X-Greylist: delayed 01:01:16.39754 by SQLgrey- Received: from frank.wiles.org (frank.wiles.org [24.124.39.75]) by svr1.postgresql.org (Postfix) with ESMTP id 97385DAC10 for ; Wed, 9 Nov 2005 20:54:35 -0400 (AST) Received: from kungfu (frank.wiles.org [127.0.0.1]) by frank.wiles.org (8.13.1/8.13.1) with SMTP id jAA0sXX3007180; Wed, 9 Nov 2005 18:54:33 -0600 Date: Wed, 9 Nov 2005 18:54:50 -0600 From: Frank Wiles To: Alvaro Herrera Cc: rjpeace@earthlink.net, yves.vindevogel@implements.be, pgsql-performance@postgresql.org Subject: Re: Some help on buffers and other performance tricks Message-Id: <20051109185450.2f20275d.frank@wiles.org> In-Reply-To: <20051110004332.GE8230@surnet.cl> References: <17035565.1131571489652.JavaMail.root@elwamui-hound.atl.sa.earthlink.net> <20051109230752.GC8230@surnet.cl> <20051109175332.3edbd5fc.frank@wiles.org> <20051110004332.GE8230@surnet.cl> X-Mailer: Sylpheed version 2.0.1 (GTK+ 2.6.10; i686-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, score=0.016 required=5 tests=[AWL=0.016] X-Spam-Score: 0.016 X-Spam-Level: X-Archive-Number: 200511/169 X-Sequence-Number: 15426 On Wed, 9 Nov 2005 21:43:33 -0300 Alvaro Herrera wrote: > Frank Wiles wrote: > > > Obviously there are systems/datasets/quantities where this won't > > always work out best, but for the majority of systems out there > > complicating your schema, maxing your hardware, and THEN tuning > > is IMHO the wrong approach. > > I wasn't suggesting to complicate the schema -- I was actually > thinking in systems where some queries are not using indexes, some > queries are plain wrong, etc. Buying a very expensive RAID and then > noticing that you just needed to create an index, is going to make > somebody feel at least somewhat stupid. Sorry I was referring to Ron statement that the first step should be to "Optimize your schema to be as tight as possible." But I agree, finding out you need an index after spending $$$ on extra hardware would be bad. Especially if you have to explain it to the person forking over the $$$! :) --------------------------------- Frank Wiles http://www.wiles.org ---------------------------------