Re: [PERFORM] Hardware/OS recommendations for large databases (</TIT= LE> </HEAD> <BODY> Mike,= On 11/15/05 6:55 AM, "Michael Stone" <mstone+postgres@mathom.u= s> wrote: <BLOCKQUOTE>On Tue, Nov 15, 2005 at 09:33:25AM -0500, Luke Lonergan= wrote: >write performance is now up to par with the best cards I believe. = ;We >find that you still need to set Linux readahead to at least 8MB >(blockdev --setra) to get maximum read performance on them, is that you= r What on earth does that do to your seek performance? </BLOCKQUOTE> We’re in decision support, as is our poster here, so seek isn’t= the issue, it’s sustained sequential transfer rate that we need. &nbs= p;At 8MB, I’d not expect too much damage though – the default is= 1.5MB. - Luke <BLOCKQUOTE> </BLOCKQUOTE> </BODY> </HTML> --B_3214896479

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 779FDDB8BA for ; Tue, 15 Nov 2005 14:48:20 -0400 (AST) 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 74252-04 for ; Tue, 15 Nov 2005 18:48:19 +0000 (GMT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from mail.mi8.com (d01gw02.mi8.com [63.240.6.46]) by svr1.postgresql.org (Postfix) with ESMTP id 12DB1DB8AE for ; Tue, 15 Nov 2005 14:48:16 -0400 (AST) Received: from 172.16.1.112 by mail.mi8.com with ESMTP (- Welcome to Mi8 Corporation www.Mi8.com (D2)); Tue, 15 Nov 2005 13:48:09 -0500 X-Server-Uuid: 7829E76E-BB9E-4995-8473-3C0929DF7DD1 Received: from MI8NYCMAIL06.Mi8.com ([172.16.1.175]) by D01SMTP02.Mi8.com with Microsoft SMTPSVC(6.0.3790.211); Tue, 15 Nov 2005 13:48:00 -0500 Received: from 67.103.45.218 ([67.103.45.218]) by MI8NYCMAIL06.Mi8.com ( [172.16.1.219]) via Exchange Front-End Server mi8owa.mi8.com ( [172.16.1.105]) with Microsoft Exchange Server HTTP-DAV ; Tue, 15 Nov 2005 13:47:59 -0500 User-Agent: Microsoft-Entourage/11.2.1.051004 Date: Tue, 15 Nov 2005 10:47:59 -0800 Subject: Re: Hardware/OS recommendations for large databases ( From: "Luke Lonergan" To: "Michael Stone" cc: pgsql-performance@postgresql.org Message-ID: Thread-Topic: [PERFORM] Hardware/OS recommendations for large databases ( Thread-Index: AcXp9Rz/g4t+KwF1QzeAxs8MLl6yOQAH/xTB In-Reply-To: <20051115145545.GQ9905@mathom.us> MIME-Version: 1.0 X-OriginalArrivalTime: 15 Nov 2005 18:48:00.0510 (UTC) FILETIME=[1A3941E0:01C5EA15] X-WSS-ID: 6F64F2ED2RS6959392-17-01 Content-Type: multipart/alternative; boundary=B_3214896479_20817725 X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, score=1.254 required=5 tests=[HTML_MESSAGE=0.001, RCVD_NUMERIC_HELO=1.253] X-Spam-Score: 1.254 X-Spam-Level: * X-Archive-Number: 200511/255 X-Sequence-Number: 15512 > This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. --B_3214896479_20817725 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Mike,=20 On 11/15/05 6:55 AM, "Michael Stone" wrote: > On Tue, Nov 15, 2005 at 09:33:25AM -0500, Luke Lonergan wrote: >> >write performance is now up to par with the best cards I believe. We >> >find that you still need to set Linux readahead to at least 8MB >> >(blockdev --setra) to get maximum read performance on them, is that you= r >=20 > What on earth does that do to your seek performance? We=B9re in decision support, as is our poster here, so seek isn=B9t the issue, it=B9s sustained sequential transfer rate that we need. At 8MB, I=B9d not expect too much damage though =AD the default is 1.5MB. - Luke >=20 --B_3214896479_20817725 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Re: [PERFORM] Hardware/OS recommendations for large databases (</TIT= LE> </HEAD> <BODY> Mike,= On 11/15/05 6:55 AM, "Michael Stone" <mstone+postgres@mathom.u= s> wrote: <BLOCKQUOTE>On Tue, Nov 15, 2005 at 09:33:25AM -0500, Luke Lonergan= wrote: >write performance is now up to par with the best cards I believe. = ;We >find that you still need to set Linux readahead to at least 8MB >(blockdev --setra) to get maximum read performance on them, is that you= r What on earth does that do to your seek performance? </BLOCKQUOTE> We’re in decision support, as is our poster here, so seek isn’t= the issue, it’s sustained sequential transfer rate that we need. &nbs= p;At 8MB, I’d not expect too much damage though – the default is= 1.5MB. - Luke <BLOCKQUOTE> </BLOCKQUOTE> </BODY> </HTML> --B_3214896479_20817725--