X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (av.hub.org [200.46.204.144]) by postgresql.org (Postfix) with ESMTP id 0327811F6B4E for ; Fri, 7 Apr 2006 22:26:21 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 36738-04 for ; Fri, 7 Apr 2006 22:26:14 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from fon.nation-net.com (213-2-2-26.uk.vianw.net [213.2.2.26]) by postgresql.org (Postfix) with ESMTP id 077F311F6912 for ; Fri, 7 Apr 2006 22:26:17 -0300 (ADT) Received: from fon.nation-net.com (localhost [127.0.0.1]) by fon.nation-net.com (Postfix) with ESMTP id 958E3137594 for ; Sat, 8 Apr 2006 02:26:17 +0100 (BST) Received: from eddie.acentral.co.uk (80-192-144-33.cable.ubr04.pres.blueyonder.co.uk [80.192.144.33]) by fon.nation-net.com (Postfix) with ESMTP id 7AE90136CD9 for ; Sat, 8 Apr 2006 02:26:17 +0100 (BST) Received: from eddie.acentral.co.uk (eddie.acentral.co.uk [127.0.0.1]) by eddie.acentral.co.uk (Postfix) with ESMTP id 7FC2175FE5 for ; Sat, 8 Apr 2006 02:26:20 +0100 (BST) Received: from localhost.localdomain (unknown [10.0.0.29]) by eddie.acentral.co.uk (Postfix) with SMTP id 57F0875F8D for ; Sat, 8 Apr 2006 02:26:20 +0100 (BST) Date: Sat, 8 Apr 2006 02:26:12 +0100 From: Gavin Hamill To: pgsql-performance@postgresql.org Subject: Re: pg 8.1.3, AIX, huge box, painfully slow. Message-Id: <20060408022612.f5918e25.gdh@laterooms.com> In-Reply-To: References: <20060407232759.45581362.gdh@laterooms.com> X-Mailer: Sylpheed version 2.1.1 (GTK+ 2.8.6; i486-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Virus-Scanned: ClamAV using ClamSMTP X-AV-Checked: ClamAV using ClamSMTP X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, score=0.106 required=5 tests=[AWL=0.106] X-Spam-Score: 0.106 X-Spam-Level: X-Archive-Number: 200604/169 X-Sequence-Number: 18253 On Fri, 07 Apr 2006 15:56:52 -0700 "Luke Lonergan" wrote: > Seems like you're hitting a very small target in RAM with these semop > calls. I wonder what part of the code is doing this - Tom would know > better how to trace it, but the equivalent of oprofile output would > be nice. I'm happy to test whatever I can, but I simply don't know enough AIX to be able to tell whether a similar kernel-level profiler is available/possible. > The other thing that I'd like to see is an evaluation of the memory > access latency of this machine from Register to RAM. I couldn't find > a benchmarking tool that was UNIX friendly out there, maybe I'll > write one real quick. I suspect this machine has a heinous latency > and a storm of semops to the same spot of RAM might be a far worse > performance problem on this machine than on others... Well, as I said to Tom, the machine is available for running tests on :) If it helps us, and helps pg become more AIX friendly, then I'm all for whatever needs done... Cheers, Gavin.