X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (av.hub.org [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 3BF2BD8C15 for ; Mon, 12 Sep 2005 18:37:32 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 15162-07 for ; Mon, 12 Sep 2005 21:37:25 +0000 (GMT) Received: from floppy.pyrenet.fr (floppy.pyrenet.fr [194.116.145.2]) by svr1.postgresql.org (Postfix) with ESMTP id EBC6ED8C53 for ; Mon, 12 Sep 2005 18:37:26 -0300 (ADT) Received: by floppy.pyrenet.fr (Postfix, from userid 106) id CFC6631F81; Mon, 12 Sep 2005 23:37:28 +0200 (MET DST) From: "Qingqing Zhou" X-Newsgroups: pgsql.performance Subject: Re: Performance considerations for very heavy INSERT traffic Date: Mon, 12 Sep 2005 17:37:19 -0700 Organization: Hub.Org Networking Services Lines: 76 Message-ID: References: <84621a6050912140464c380c3@mail.gmail.com> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0025_01C5B7C0.A035C750" X-Complaints-To: usenet@news.hub.org X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 To: pgsql-performance@postgresql.org X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=1.243 required=5 tests=[AWL=-0.038, HTML_50_60=0.095, HTML_MESSAGE=0.001, PRIORITY_NO_NAME=1.185] X-Spam-Level: * X-Archive-Number: 200509/148 X-Sequence-Number: 14456 This is a multi-part message in MIME format. ------=_NextPart_000_0025_01C5B7C0.A035C750 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable "Brandon Black" wrote ... Increasing shared_buffers seems to always help, even out to half of = the dev box's ram (2G). =20 Though officially PG does not prefer huge shared_buffers size, I did = see several times that performance was boosted in case IO is the = bottleneck. Also, if you want to use big bufferpool setting, make sure = your version has Tom's split BufMgrLock patch = (http://archives.postgresql.org/pgsql-committers/2005-03/msg00025.php), = which might already in 8.0.x somewhere. And if you want to use = bufferpool bigger than 2G on 64-bit machine, you may need 8.1 = (http://archives.postgresql.org/pgsql-committers/2005-08/msg00221.php). Regards, Qingqing ------=_NextPart_000_0025_01C5B7C0.A035C750 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
 
"Brandon Black" <blblack@gmail.com> wrote = ...

Increasing shared_buffers seems to always help, even out to = half of=20 the dev box's ram (2G). 
 
Though officially PG does not prefer = huge=20 shared_buffers size, I did see several times = that performance was=20 boosted in case IO is the bottleneck. Also, if you want to use big = bufferpool=20 setting, make sure your version has Tom's split BufMgrLock = patch
(http://archives.postgresql.org/pgsql-committers/2005-03/msg00025.php= ),=20 which might already in 8.0.x somewhere. And if you want to use = bufferpool=20 bigger than 2G on 64-bit machine, you may need 8.1 (http://archives.postgresql.org/pgsql-committers/2005-08/msg00221.php= ).
 
Regards,
Qingqing
------=_NextPart_000_0025_01C5B7C0.A035C750--