Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 6EFF2475D12 for ; Thu, 3 Oct 2002 11:49:43 -0400 (EDT) Received: from www.pspl.co.in (www.pspl.co.in [202.54.11.65]) by postgresql.org (Postfix) with ESMTP id EFE17475C15 for ; Thu, 3 Oct 2002 11:49:41 -0400 (EDT) Received: (from root@localhost) by www.pspl.co.in (8.11.6/8.11.6) id g93FpDP03689 for ; Thu, 3 Oct 2002 21:21:13 +0530 Received: from daithan (daithan.intranet.pspl.co.in [192.168.7.161]) by www.pspl.co.in (8.11.6/8.11.0) with ESMTP id g93FpDv03684 for ; Thu, 3 Oct 2002 21:21:13 +0530 From: "Shridhar Daithankar" To: pgsql-general Date: Thu, 03 Oct 2002 21:20:16 +0530 MIME-Version: 1.0 Subject: Re: Large databases, performance Reply-To: shridhar_daithankar@persistent.co.in Message-ID: <3D9CB490.2412.A78039E@localhost> In-reply-to: <3D9C53B3.2050508@clearmetrix.com> X-mailer: Pegasus Mail for Windows (v4.02) Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Content-description: Mail message body X-Virus-Scanned: by AMaViS new-20020517 X-Archive-Number: 200210/125 X-Sequence-Number: 31124 On 3 Oct 2002 at 10:26, Charles H. Woloszynski wrote: > Forgive my ignorance, but what about 2.4.19-16 is that much faster? Are > we talking about 2x improvement for your tests? We are currently on > 2.4.9 and looking at the performance and wondering... so any comments > are appreciated. Well, for one thing, 2.4.19 contains backported O(1) scheduler patch which improves SMP performance by heaps as task queue is per cpu rather than one per system. I don't think any system routinely runs thousands of processes unless it's a web/ftp/mail server. In that case improved scheduling wuld help as well.. Besides there were major VM rewrites/changes after 2.4.10 which corrected almost all the major VM fiaskos on linux. For anything VM intensive it's recommended that you run 2.4.17 at least. I would say it's worth going for it. Bye Shridhar -- Sturgeon's Law: 90% of everything is crud.