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 A2032D1E441 for ; Wed, 17 Mar 2004 05:14:27 +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 60910-08 for ; Wed, 17 Mar 2004 01:14:26 -0400 (AST) Received: from outbound.mailhop.org (outbound.mailhop.org [63.208.196.171]) by svr1.postgresql.org (Postfix) with ESMTP id 9EE13D1E90B for ; Wed, 17 Mar 2004 01:14:25 -0400 (AST) Received: from ool-4352919e.dyn.optonline.net ([67.82.145.158] helo=zeut.net) by outbound.mailhop.org with asmtp (TLSv1:AES256-SHA:256) (Exim 4.20) id 1B3TNm-000PrR-Be; Wed, 17 Mar 2004 00:14:22 -0500 Message-ID: <4057DE34.1080001@zeut.net> Date: Wed, 17 Mar 2004 00:12:20 -0500 From: "Matthew T. O'Connor" User-Agent: Mozilla Thunderbird 0.5 (Windows/20040207) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Joe Conway Cc: Tom Lane , "Matthew T. O'Connor" , pgsql-performance , Josh Berkus Subject: Re: rapid degradation after postmaster restart References: <4052661D.5060505@joeconway.com> <14884.1079150555@sss.pgh.pa.us> <4055F343.6070600@joeconway.com> <405614CA.5030101@zeut.net> <405689A4.4080406@joeconway.com> <16735.1079414723@sss.pgh.pa.us> <4057D8BD.4080902@joeconway.com> In-Reply-To: <4057D8BD.4080902@joeconway.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Mail-Handler: MailHop Outbound by DynDNS.org X-Report-Abuse-To: abuse@dyndns.org X-MHO-User: Zeut X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200403/231 X-Sequence-Number: 6087 On Tue, 2004-03-16 at 23:49, Joe Conway wrote: > I have tested Tom's original patch now. The good news -- it works great > in terms of reducing the load imposed by vacuum -- almost to the level > of being unnoticeable. The bad news -- in a simulation test which loads > an hour's worth of data, even with delay set to 1 ms, vacuum of the > large table exceeds two hours (vs 12-14 minutes with delay = 0). Since > that hourly load is expected 7 x 24, this obviously isn't going to work. If memory serves, the problem is that you actually sleep 10ms even when you set it to 1. One of the thing changed in Jan's later patch was the ability to specify how many pages to work on before sleeping, rather than how long to sleep inbetween every 1 page. You might be able to do a quick hack and have it do 10 pages or so before sleeping. Matthew