Received: from localhost (unknown [200.46.208.211]) by mail.postgresql.org (Postfix) with ESMTP id 428466346DB for ; Sat, 14 Mar 2009 01:03:00 -0300 (ADT) Received: from mail.postgresql.org ([200.46.204.86]) by localhost (mx1.hub.org [200.46.208.211]) (amavisd-maia, port 10024) with ESMTP id 79325-06 for ; Sat, 14 Mar 2009 01:02:52 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from exprod7og106.obsmtp.com (exprod7og106.obsmtp.com [64.18.2.165]) by mail.postgresql.org (Postfix) with SMTP id CF0EF634669 for ; Sat, 14 Mar 2009 01:02:19 -0300 (ADT) Received: from source ([64.233.182.185]) by exprod7ob106.postini.com ([64.18.6.12]) with SMTP ID DSNKSbssSv8kLbRLeBw79n/L+UfMxbDZWTwk@postini.com; Fri, 13 Mar 2009 21:02:22 PDT Received: by nf-out-0910.google.com with SMTP id 30so1550463nfu.27 for ; Fri, 13 Mar 2009 21:02:18 -0700 (PDT) Received: by 10.210.65.15 with SMTP id n15mr1352679eba.19.1237003338274; Fri, 13 Mar 2009 21:02:18 -0700 (PDT) Received: from oxford.xeocode.com.enterprisedb.com ([87.127.95.198]) by mx.google.com with ESMTPS id 5sm4605478nfv.61.2009.03.13.21.02.17 (version=TLSv1/SSLv3 cipher=RC4-MD5); Fri, 13 Mar 2009 21:02:17 -0700 (PDT) To: Robert Haas Cc: Tom Lane , Alan Stange , "Jignesh K. Shah" , Greg Smith , "pgsql-performance\@postgresql.org" Subject: Re: 8.4 Performance improvements: was Re: Proposal of tunable fix for scalability of 8.4 In-Reply-To: <603c8f070903131937t1c33fa97k29e7d8d07fa2e83@mail.gmail.com> (Robert Haas's message of "Fri\, 13 Mar 2009 22\:37\:37 -0400") User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.1 (gnu/linux) X-Draft-From: ("nnimap+imap.gmail.com:PG" 113050) References: <49B922C0.6050700@sun.com> <49BA5537.8060002@sun.com> <87prgl4jbh.fsf@oxford.xeocode.com> <49BA79A4.9030208@rentec.com> <12431.1236965004@sss.pgh.pa.us> <87d4ck50h1.fsf@oxford.xeocode.com> <26666.1236996376@sss.pgh.pa.us> <603c8f070903131937t1c33fa97k29e7d8d07fa2e83@mail.gmail.com> From: Gregory Stark Organization: EnterpriseDB Date: Sat, 14 Mar 2009 04:02:15 +0000 Message-ID: <878wn84urc.fsf@oxford.xeocode.com> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200903/184 X-Sequence-Number: 33106 Robert Haas writes: > On Fri, Mar 13, 2009 at 10:06 PM, Tom Lane wrote: > >> I assume you meant effective_io_concurrency. =A0We'd still need a special >> case because the default is currently hard-wired at 1, not 0, if >> configure thinks the function exists. =A0Also there's a posix_fadvise ca= ll >> in xlog.c that that parameter doesn't control anyhow. > > I think 1 should mean no prefetching, rather than 0. If the number of > concurrent I/O requests was 0, that would mean you couldn't perform > any I/O at all. That is actually how I had intended it but apparently I messed it up at some point such that later patches were doing some prefetching at 1 and there was no way to disable it. When Tom reviewed it he corrected the inability to disable prefetching by making 0 disable prefetching. I didn't think it was worth raising as an issue but I didn't realize we were currently doing prefetching by default? i didn't realize that. Even on a system with posix_fadvise there's nothing much to be gained unless the data= is on a RAID device, so the original objection holds anyways. We shouldn't do = any prefetching unless the user tells us to. --=20 Gregory Stark EnterpriseDB http://www.enterprisedb.com Ask me about EnterpriseDB's 24x7 Postgres support!