X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.2]) by svr1.postgresql.org (Postfix) with ESMTP id 3489FD1E149 for ; Tue, 6 Apr 2004 12:52:45 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (neptune.hub.org [200.46.204.2]) (amavisd-new, port 10024) with ESMTP id 07714-07 for ; Tue, 6 Apr 2004 12:52:43 -0300 (ADT) Received: from flake.decibel.org (flake.decibel.org [66.143.173.58]) by svr1.postgresql.org (Postfix) with SMTP id 5BA0CD1DC17 for ; Tue, 6 Apr 2004 12:52:42 -0300 (ADT) Received: (qmail 737 invoked by uid 1001); 6 Apr 2004 15:52:35 -0000 Date: Tue, 6 Apr 2004 10:52:35 -0500 From: "Jim C. Nasby" To: Tom Lane Cc: Qing Zhao , pgsql-performance@postgresql.org Subject: Re: possible improvement between G4 and G5 Message-ID: <20040406155235.GZ74840@nasby.net> References: <009901c41b48$8357c860$2500fa0a@CardServices.TCI.com> <19781.1081230442@sss.pgh.pa.us> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <19781.1081230442@sss.pgh.pa.us> User-Agent: Mutt/1.4.2.1i X-Operating-System: FreeBSD 4.9-RELEASE-p3 i386 X-Distributed: Join the Effort! http://www.distributed.net X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200404/65 X-Sequence-Number: 6365 On Tue, Apr 06, 2004 at 01:47:22AM -0400, Tom Lane wrote: > Qing Zhao writes: > > We have got a G5 64-bit processor to replace an old G4 32-bit > > processor. Given everything else equal, should we see a big > > improvement on PG's performance? > > Nope. Database performance typically depends on disk performance first, > and RAM size second. A 64-bit processor might help by allowing you to > install more RAM, but you didn't say that you had. Memory bandwidth is a consideration too, so you might see some performance improvements on a G5. We recently debated between Xeons and Opterons in a new PGSQL server and a little poking around on the lists indicated that the Opterons did perform better, presumably due to the increased memory bandwidth. Incidentally, this is why you need about 2x the CPUs on Sun hardware vs RS6000 hardware for database stuff (and that gets expensive if you're paying per CPU!). -- Jim C. Nasby, Database Consultant jim@nasby.net Member: Triangle Fraternity, Sports Car Club of America Give your computer some brain candy! www.distributed.net Team #1828 Windows: "Where do you want to go today?" Linux: "Where do you want to go tomorrow?" FreeBSD: "Are you guys coming, or what?"