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 5B4A9D1B50E for ; Thu, 13 Nov 2003 15:52:58 +0000 (GMT) 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 63479-03 for ; Thu, 13 Nov 2003 11:52:31 -0400 (AST) Received: from bramble.mmrd.com (unknown [65.217.53.66]) by svr1.postgresql.org (Postfix) with ESMTP id 5995ED1B527 for ; Thu, 13 Nov 2003 11:52:26 -0400 (AST) Received: from thorn.mmrd.com (thorn.mmrd.com [172.25.10.100]) by bramble.mmrd.com (8.12.8/8.12.8) with ESMTP id hADEvOcM008935; Thu, 13 Nov 2003 09:57:24 -0500 Received: from gnvex001.mmrd.com (gnvex001.mmrd.com [192.168.3.55]) by thorn.mmrd.com (8.11.6/8.11.6) with ESMTP id hADFqPl26573; Thu, 13 Nov 2003 10:52:25 -0500 Received: from camel.mmrd.com ([172.25.5.213]) by gnvex001.mmrd.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id V3JRTPZH; Thu, 13 Nov 2003 10:52:24 -0500 Subject: Re: Value of Quad vs. Dual Processor machine From: Robert Treat To: Jeff Cc: Chris Field , ron.l.johnson@cox.net, pgsql-performance@postgresql.org In-Reply-To: <20031112092807.14b18bf4.threshar@torgo.978.org> References: <1068593567.2063.8.camel@tuxsrva> <1068600291.30969.53.camel@haggis.homelan> <001401c3a8c2$8c8fec30$040210ac@win2000> <20031112092807.14b18bf4.threshar@torgo.978.org> Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Ximian Evolution 1.0.8 Date: 13 Nov 2003 10:52:25 -0500 Message-Id: <1068738745.10945.18380.camel@camel> Mime-Version: 1.0 X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200311/87 X-Sequence-Number: 4647 On Wed, 2003-11-12 at 09:28, Jeff wrote: > On Tue, 11 Nov 2003 21:13:19 -0500 > "Chris Field" wrote: > > > we are looking at Xeon, We are currently running it on a quad sun v880 > > compiled to be 64bit and have been getting dreadful performance. I > > don't think we really have much to gain from going 64bit. > > > > > By chance, are you running 7.3.4 on that sun? > If so, try this: > export CFLAGS=-02 > ./configure > > and rebuild PG. > > Before 7.4 PG was build with _no_ optimization on Solaris. > Recompiling gives __HUGE__ (notice the underscores) performance gains. > > And onto the dual vs quad. > > PG will only use 1 cpu / connection / query. > > So if your machine iwll have 1-2 queries running at a time those other 2 > proc's will sit around idling. However if you are going to have a bunch > going, 4 cpus will be most useful. One of hte nicest things to do for > PG is more ram and fast IO. It really loves those things. > We've just started kicking around the idea of moving one of our boxes to a quad-proc machine from a dual. Under normal circumstances the 2 processors handle maybe 200 transactions per second with 90% system idle. However we have people who occasionally run historical reports on our data, and those reports are fairly CPU intensive. Usually it is not a problem for the main web system, but when pg_dump is running, that is also cpu intensive, so we end up with two highly cpu intensive items running on our machine, and we start to notice issues on the main web system. Robert Treat -- Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL