Received: from localhost (maia-5.hub.org [200.46.204.182]) by postgresql.org (Postfix) with ESMTP id 197539FB468 for ; Thu, 1 Mar 2007 18:28:33 -0400 (AST) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.182]) (amavisd-new, port 10024) with ESMTP id 36029-05 for ; Thu, 1 Mar 2007 18:28:24 -0400 (AST) X-Greylist: from auto-whitelisted by SQLgrey-1.7.4 Received: from floppy.pyrenet.fr (news.pyrenet.fr [194.116.145.2]) by postgresql.org (Postfix) with ESMTP id 3555E9FB39D for ; Thu, 1 Mar 2007 18:28:27 -0400 (AST) Received: by floppy.pyrenet.fr (Postfix, from userid 106) id E04BF308E7; Thu, 1 Mar 2007 23:28:24 +0100 (MET) From: "hatman" X-Newsgroups: pgsql.performance Subject: Re: performances with Pentium D Date: 1 Mar 2007 14:28:09 -0800 Organization: http://groups.google.com Lines: 50 Message-ID: <1172788089.781005.235200@30g2000cwc.googlegroups.com> References: <1172678294.544317.195100@t69g2000cwt.googlegroups.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Complaints-To: groups-abuse@google.com Cc: ben.trewern@mowlem.com In-Reply-To: User-Agent: G2/1.0 X-HTTP-UserAgent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.12) Gecko/20060607 Debian/1.7.12-1.2,gzip(gfe),gzip(gfe) Complaints-To: groups-abuse@google.com Injection-Info: 30g2000cwc.googlegroups.com; posting-host=213.103.157.52; posting-account=gnwH_QwAAAAQLNv14JlAW4Wgi-SQgnth To: pgsql-performance@postgresql.org X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0.001 tagged_above=0 required=5 tests=BAYES_50 X-Spam-Level: X-Archive-Number: 200703/106 X-Sequence-Number: 23511 Hi Ben, Thanks for you answer. I was thinking about CPU speed bottleneck because one of the CPU load was always quite low on the server (Pentium D, dual core) and on the other hand, the CPU load on my laptop was always very high. After some more testing (using a threaded client software which does the same inserts using 10 parallel connections), i was able to have the other CPU arround 90% too :-) Then the INSERT speed reached 18'181 inserts/sec !! So now i'm wondering if i reached the disk limit or again the CPU limit... Anyway, thanks a lot for your advice, i didn't know this difference between fsync implementation on SCSI and IDE ! Jo=EBl On Mar 1, 12:00 pm, "Ben Trewern" wrote: > I suspect the difference is your disk subsystem. IDE disks (in your lapt= op > I assume) quite often (almost always!!) ignore fsync calls and return as > soon as the data gets to the disk cache, not the physical disk. SCSI dis= ks > are almost always more correct, and wait until the data gets to the physi= cal > disk before they return from an fsync call. > > I hope this helps. > > Regards, > > Ben"hatman" wrote in message > > news:1172678294.544317.195100@t69g2000cwt.googlegroups.com... > Hi all, > > Do someone already had some problem with performances using a pentium > D (64 bits) and postgres 8.2.3 on a redhat enterprise update 2 ? > I did the install from sources and nothing change... I have a RAID 0 > for data and 3Gb of RAM and my inserts rate is quite low, 8333 inserts/ > sec (lower than on my laptop which is 10526 inserts/sec). > I suspect a problem with the CPU because using gkrellm, the use of 1 > CPU is always quite low... Is it normal ? > > Many thanks for your help, > > Jo=EBl