Re: [PERFORM] Hardware/OS recommendations for large databases (</TIT= LE> </HEAD> <BODY> Vivek= , On 11/18/05 8:07 AM, "Vivek Khera" <vivek@khera.org> wrote:= <BLOCKQUOTE> On Nov 18, 2005, at 10:13 AM, Luke Lonergan wrote: <BLOCKQUOTE>Still, there is a CPU limit here – this is not I/= O bound, it is CPU limited as evidenced by the sensitivity to readahead sett= ings. =A0=A0If the filesystem could do 1GB/s, you wouldn’t go any faster t= han 244MB/s. </BLOCKQUOTE> Yeah, and mysql would probably be faster on your trivial queries.=A0 Try conc= urrent large joins and updates and see which system is faster. </BLOCKQUOTE> That’s what we do to make a living. And it’s Oracle that = a lot faster because they implemented a much tighter, optimized I/O p= ath to disk than Postgres. Since you asked, we bought the 5 systems as a cluster – and with Bizg= res MPP we get close to 400MB/s per machine on complex queries. - Luke </BODY> </HTML> --B_3215146600

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 D9562D82C0 for ; Fri, 18 Nov 2005 12:16:57 -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 53168-08 for ; Fri, 18 Nov 2005 16:16:55 +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 BC3D8DBAC0 for ; Fri, 18 Nov 2005 12:16:53 -0400 (AST) Received: from 172.16.1.110 by mail.mi8.com with ESMTP (- Welcome to Mi8 Corporation www.Mi8.com (D2)); Fri, 18 Nov 2005 11:16:47 -0500 X-Server-Uuid: 7829E76E-BB9E-4995-8473-3C0929DF7DD1 Received: from MI8NYCMAIL06.Mi8.com ([172.16.1.175]) by D01SMTP01.Mi8.com with Microsoft SMTPSVC(6.0.3790.211); Fri, 18 Nov 2005 11:16:40 -0500 Received: from 69.181.100.71 ([69.181.100.71]) by MI8NYCMAIL06.Mi8.com ( [172.16.1.219]) via Exchange Front-End Server mi8owa.mi8.com ( [172.16.1.104]) with Microsoft Exchange Server HTTP-DAV ; Fri, 18 Nov 2005 11:16:40 -0500 User-Agent: Microsoft-Entourage/11.2.1.051004 Date: Fri, 18 Nov 2005 08:16:39 -0800 Subject: Re: Hardware/OS recommendations for large databases ( From: "Luke Lonergan" To: "Vivek Khera" , "Postgresql Performance" Message-ID: Thread-Topic: [PERFORM] Hardware/OS recommendations for large databases ( Thread-Index: AcXsWvFEknbFGdXQQ46m0ym7HfvpqgAAIMyI In-Reply-To: <1F5BD2B1-8A75-498E-ADB7-8CB661066013@khera.org> MIME-Version: 1.0 X-OriginalArrivalTime: 18 Nov 2005 16:16:40.0865 (UTC) FILETIME=[7592A110:01C5EC5B] X-WSS-ID: 6F6321E52RS9512144-08-01 Content-Type: multipart/alternative; boundary=B_3215146600_2400117 X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, score=1.254 required=5 tests=[AWL=-0.000, HTML_MESSAGE=0.001, RCVD_NUMERIC_HELO=1.253] X-Spam-Score: 1.254 X-Spam-Level: * X-Archive-Number: 200511/374 X-Sequence-Number: 15631 > 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_3215146600_2400117 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Vivek, On 11/18/05 8:07 AM, "Vivek Khera" wrote: >=20 > On Nov 18, 2005, at 10:13 AM, Luke Lonergan wrote: >=20 >> Still, there is a CPU limit here =AD this is not I/O bound, it is CPU limi= ted >> as evidenced by the sensitivity to readahead settings. =A0=A0If the filesyst= em >> could do 1GB/s, you wouldn=B9t go any faster than 244MB/s. >=20 > Yeah, and mysql would probably be faster on your trivial queries.=A0 Try > concurrent large joins and updates and see which system is faster. That=B9s what we do to make a living. And it=B9s Oracle that a lot faster because they implemented a much tighter, optimized I/O path to disk than Postgres. Since you asked, we bought the 5 systems as a cluster =AD and with Bizgres MP= P we get close to 400MB/s per machine on complex queries. - Luke --B_3215146600_2400117 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> Vivek= , On 11/18/05 8:07 AM, "Vivek Khera" <vivek@khera.org> wrote:= <BLOCKQUOTE> On Nov 18, 2005, at 10:13 AM, Luke Lonergan wrote: <BLOCKQUOTE>Still, there is a CPU limit here – this is not I/= O bound, it is CPU limited as evidenced by the sensitivity to readahead sett= ings. =A0=A0If the filesystem could do 1GB/s, you wouldn’t go any faster t= han 244MB/s. </BLOCKQUOTE> Yeah, and mysql would probably be faster on your trivial queries.=A0 Try conc= urrent large joins and updates and see which system is faster. </BLOCKQUOTE> That’s what we do to make a living. And it’s Oracle that = a lot faster because they implemented a much tighter, optimized I/O p= ath to disk than Postgres. Since you asked, we bought the 5 systems as a cluster – and with Bizg= res MPP we get close to 400MB/s per machine on complex queries. - Luke </BODY> </HTML> --B_3215146600_2400117--