X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id EBB8B56899 for ; Tue, 22 Mar 2005 21:03:49 +0000 (GMT) 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 77851-08 for ; Tue, 22 Mar 2005 21:03:46 +0000 (GMT) Received: from smtp-gw-cl-c.dmv.com (smtp-gw-cl-c.dmv.com [216.240.97.41]) by svr1.postgresql.org (Postfix) with ESMTP id 63A9F568D8 for ; Tue, 22 Mar 2005 21:03:45 +0000 (GMT) Received: from lanshark.dmv.com (lanshark.dmv.com [216.240.97.46]) by smtp-gw-cl-c.dmv.com (8.12.10/8.12.10) with ESMTP id j2MLMoEo019689; Tue, 22 Mar 2005 16:22:50 -0500 (EST) (envelope-from sven@dmv.com) Subject: Re: PostgreSQL on Solaris 8 and ufs From: Sven Willenberger To: Brandon Metcalf Cc: pgsql-performance@postgresql.org In-Reply-To: References: Content-Type: text/plain Date: Tue, 22 Mar 2005 16:03:43 -0500 Message-Id: <1111525423.9054.16.camel@lanshark.dmv.com> Mime-Version: 1.0 X-Mailer: Evolution 2.0.4 Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.39 X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0.335 tagged_above=0 required=5 tests=AWL, DNS_FROM_RFC_WHOIS X-Spam-Level: X-Archive-Number: 200503/469 X-Sequence-Number: 11174 On Tue, 2005-03-22 at 14:44 -0600, Brandon Metcalf wrote: > We've recently moved our pgsql installation and DBs to a Solaris 8 > machine with striped and mirrored ufs filesystem that houses the DB > data. We are now seeing terrible performance and the bottleneck is no > doubt disk I/O. > > We've tried modifying a tunables related to ufs, but it doesn't seem > to be helping. > > Is there anything we should be looking at that is specifically related > to ufs filesystems on Solaris 8 or possibly something in general that > would improve performance? > > Thanks. > What are you using to create your raid? You say it is "no doubt disk I/O" - does iostat confirm this? A lot of performance issues are related to the size of the stripe you chose for the striped portion of the array, the actual array configuration, etc. I am assuming you have looked at system variables such as autoup and the likes? What tweaks have you done? Also, are your pg_xlog and data directories separated onto separate volumes? Doing so will help immensely. What are you using to measure performance? Sven