Received: from maia.hub.org (maia-2.hub.org [200.46.204.251]) by mail.postgresql.org (Postfix) with ESMTP id 7BF58632B76 for ; Fri, 7 May 2010 21:16:33 -0300 (ADT) Received: from mail.postgresql.org ([200.46.204.86]) by maia.hub.org (mx1.hub.org [200.46.204.251]) (amavisd-maia, port 10024) with ESMTP id 39483-04 for ; Sat, 8 May 2010 00:16:25 +0000 (UTC) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from dd1514.kasserver.com (dd1514.kasserver.com [85.13.128.107]) by mail.postgresql.org (Postfix) with ESMTP id A3E08632778 for ; Fri, 7 May 2010 21:16:26 -0300 (ADT) Received: from [192.168.1.3] (dslb-094-220-230-086.pools.arcor-ip.net [94.220.230.86]) by dd1514.kasserver.com (Postfix) with ESMTP id 955E3187408A; Sat, 8 May 2010 02:16:16 +0200 (CEST) Date: Sat, 08 May 2010 02:16:23 +0200 From: Bernd Helmle To: Tom Lane cc: Kevin Grittner , Andrew Dunstan , pgsql-hackers@postgresql.org Subject: Re: no universally correct setting for fsync Message-ID: In-Reply-To: <3860.1273276155@sss.pgh.pa.us> References: <4BE3D3930200002500031382@gw.wicourts.gov> <4BE425F7.30804@dunslane.net> <4BE3E205020000250003138F@gw.wicourts.gov> <3860.1273276155@sss.pgh.pa.us> X-Mailer: Mulberry/4.0.8 (Mac OS X) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=-0.001 tagged_above=-5 required=5 tests=BAYES_40=-0.001 X-Spam-Level: X-Archive-Number: 201005/383 X-Sequence-Number: 162046 --On 7. Mai 2010 19:49:15 -0400 Tom Lane wrote: > Bernd Helmle writes: >> I've recently even started to wonder if the performance gain with >> fsync=off is still that large on modern hardware. While testing large >> migration procedures to a new version some time ago (on an admitedly >> fast storage) i forgot here and then to turn it off, without a >> significant degradation in performance. > > That says to me either that you're using a battery-backed write cache, > or your fsyncs don't really work (no write barriers or something like > that). > Well, yes, BBU present and proven storage. Maybe i'm wrong, but it seems battery backed write caches aren't that seldom even in low end systems nowadays. -- Thanks Bernd