Received: from localhost (maia-1.hub.org [200.46.204.191]) by postgresql.org (Postfix) with ESMTP id 328499FBCFA for ; Mon, 4 Jun 2007 12:27:37 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.191]) (amavisd-maia, port 10024) with ESMTP id 28616-02 for ; Mon, 4 Jun 2007 12:27:33 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.4 Received: from nabagan.bluegap.ch (nabagan.bluegap.ch [88.198.58.248]) by postgresql.org (Postfix) with ESMTP id A6AD39FBCDB for ; Mon, 4 Jun 2007 12:27:33 -0300 (ADT) Received: from [192.168.0.237] (berlin.programmfabrik.de [::ffff:213.221.101.147]) (AUTH: CRAM-MD5 markus@bluegap.ch, TLS: TLSv1/SSLv3, 256bits, AES256-SHA) by nabagan.bluegap.ch with esmtp; Mon, 04 Jun 2007 18:21:43 +0200 id 0041DB56.46643C17.0000485F Message-ID: <46642F2E.3080509@bluegap.ch> Date: Mon, 04 Jun 2007 17:26:38 +0200 From: Markus Schiltknecht User-Agent: Icedove 1.5.0.10 (X11/20070329) MIME-Version: 1.0 To: Heikki Linnakangas CC: pgsql-performance@postgresql.org Subject: Re: dbt2 NOTPM numbers References: <466428F7.5010808@bluegap.ch> <46642B33.6030101@enterprisedb.com> In-Reply-To: <46642B33.6030101@enterprisedb.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200706/31 X-Sequence-Number: 25064 Hi, Heikki Linnakangas wrote: > There's clearly something wrong. The response times are ridiculously > high, they should be < 5 seconds (except for stock level transaction) to > pass a TPC-C test. I wonder if you built any indexes at all? Hm.. according to the output/5/db/plan0.out, all queries use index scans, so that doesn't seem to be the problem. > The configuration I'm running here has 3 data drives, and I'm getting > reasonable results with ~100 warehouses, at ~1200 noTPM. Thanks, that's exactly the one simple and very raw comparison value I've been looking for. (Since most of the results pages of (former?) OSDL are down). I'll run a bonnie++ first. As the CPUs seem to be idle most of the time (see the vmstat.out below), I'm suspecting the RAID or disks. Regards Markus procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 2 2 329512 1289384 8 2680704 0 0 4 1 1 2 0 0 100 0 0 48 329512 1733016 8 2083400 0 0 1789 2265 553 1278 1 1 63 35 2 81 329512 492052 8 3194532 0 0 6007 10135 1025 3291 6 1 21 71 0 8 329512 153360 8 3457936 0 0 6321 11919 1093 4581 7 2 12 79 0 9 329512 150188 8 3433380 0 0 2083 5078 707 2197 2 1 35 62 0 6 329512 148412 8 3408748 0 0 1001 2888 526 1203 1 0 34 64 0 27 329512 152212 8 3379736 0 0 2281 5166 733 2320 3 1 18 79 0 11 329512 152560 8 3355940 0 0 1837 4028 626 1738 2 1 35 63 0 14 329512 149268 8 3334912 0 0 1674 3836 630 1619 2 1 31 67 0 6 329512 152916 8 3311552 0 0 1404 3017 568 1372 1 0 57 41 0 13 329688 149492 8 3313200 0 0 1687 4178 650 1644 2 1 29 69 0 84 329688 153480 8 3309684 0 0 812 3790 641 2669 1 1 22 76 0 18 329688 149232 8 3314032 0 0 87 2147 511 2414 0 0 16 83 3 20 329688 149196 8 3314648 0 0 756 1854 496 1044 1 0 52 47