X-Original-To: pgsql-performance-postgresql.org@postgresql.org Received: from localhost (mx1.hub.org [200.46.208.251]) by postgresql.org (Postfix) with ESMTP id 929619FA10A for ; Mon, 12 Jun 2006 22:50:43 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.208.251]) (amavisd-new, port 10024) with ESMTP id 01888-09 for ; Mon, 12 Jun 2006 22:50:37 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey- Received: from mail.Mi8.com (d01gw01.mi8.com [63.240.6.47]) by postgresql.org (Postfix) with ESMTP id 589D59FA4A8 for ; Mon, 12 Jun 2006 22:50:39 -0300 (ADT) Received: from 172.16.1.110 by mail.Mi8.com with ESMTP (- Welcome to Mi8 Corporation www.Mi8.com (D1)); Mon, 12 Jun 2006 21:50:31 -0400 X-Server-Uuid: 241911D6-425B-44B9-A073-E3FE0F8FC774 Received: from MI8NYCMAIL06.Mi8.com ([172.16.1.160]) by D01SMTP01.Mi8.com with Microsoft SMTPSVC(6.0.3790.1830); Mon, 12 Jun 2006 21:50:31 -0400 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 ( [172.16.1.104]) with Microsoft Exchange Server HTTP-DAV ; Tue, 13 Jun 2006 01:50:30 +0000 User-Agent: Microsoft-Entourage/11.2.3.060209 Date: Mon, 12 Jun 2006 18:50:27 -0700 Subject: Re: 64-bit vs 32-bit performance ... backwards? From: "Luke Lonergan" To: "Steve Atkins" , "Pgsql-Performance ((E-mail))" Message-ID: Thread-Topic: [PERFORM] 64-bit vs 32-bit performance ... backwards? Thread-Index: AcaOi74+/MS5Dvp+EdqF7QAWy4o9DA== In-Reply-To: <1BAED8DB-01CE-4D10-A509-91FB4DF1F04D@blighty.com> MIME-Version: 1.0 X-OriginalArrivalTime: 13 Jun 2006 01:50:31.0282 (UTC) FILETIME=[C0CBD520:01C68E8B] X-WSS-ID: 6890C66D1N411290829-01-01 Content-Type: multipart/alternative; boundary=B_3232983027_1886387 X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200606/152 X-Sequence-Number: 19509 > 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_3232983027_1886387 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Opteron is ~20% faster at executing code in 64-bit mode than 32-bit because of the extra registers made available with their 64-bit mode: http://www.tomshardware.com/2003/04/22/duel_of_the_titans/page7.html Doubling the GPRs from 8 to 16 has generally made a 20%-30% difference in CPU-bound work: http://www.tomshardware.com/2003/04/22/duel_of_the_titans/page18.html If the task is memory bandwidth bound, there should be an advantage to using less memory for the same task, but if the database is using types that are the same width for either execution mode, you wouldn't expect a significant difference just from wider pointer arithmetic. - Luke --B_3232983027_1886387 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: quoted-printable Re: [PERFORM] 64-bit vs 32-bit performance ... backwards?
Opteron is ~20% faster at executing code in 64-bit mode than 32-bit because= of the extra registers made available with their 64-bit mode:
  http://www.tomshardware.com/2003/04/22/duel_of_the_titans/pa= ge7.html

Doubling the GPRs from 8 to 16 has generally made a 20%-30% difference in C= PU-bound work:
  http://www.tomshardware.com/2003/04/22/duel_of_the_titans/p= age18.html

If the task is memory bandwidth bound, there should be an advantage to usin= g less memory for the same task, but if the database is using types that are= the same width for either execution mode, you wouldn't expect a significant= difference just from wider pointer arithmetic.

- Luke --B_3232983027_1886387--