X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [64.117.224.130]) by svr1.postgresql.org (Postfix) with ESMTP id D7C5DD1B53A for ; Tue, 2 Sep 2003 16:01:57 +0000 (GMT) Received: from svr1.postgresql.org ([64.117.224.193]) by localhost (neptune.hub.org [64.117.224.130]) (amavisd-new, port 10024) with ESMTP id 76875-06 for ; Tue, 2 Sep 2003 13:01:27 -0300 (ADT) Received: from lorax.kciLink.com (lorax.kciLink.com [206.112.95.1]) by svr1.postgresql.org (Postfix) with ESMTP id 063B1D1B8A9 for ; Tue, 2 Sep 2003 13:01:27 -0300 (ADT) Received: from localhost (localhost [127.0.0.1]) by lorax.kciLink.com (Postfix) with ESMTP id 01C953ECA for ; Tue, 2 Sep 2003 12:01:27 -0400 (EDT) Received: from lorax.kciLink.com ([127.0.0.1]) by localhost (lorax.kciLink.com [127.0.0.1]) (amavisd-new, port 10024) with SMTP id 72262-01 for ; Tue, 2 Sep 2003 12:01:16 -0400 (EDT) Received: from lorax.kciLink.com (localhost [127.0.0.1]) by lorax.kciLink.com (Postfix) with ESMTP id 73B713EC4 for ; Tue, 2 Sep 2003 12:01:16 -0400 (EDT) Received: (from news@localhost) by lorax.kciLink.com (8.12.8p1/8.12.8/Submit) id h82G1GHc056157 for pgsql-performance@postgresql.org; Tue, 2 Sep 2003 12:01:16 -0400 (EDT) To: pgsql-performance@postgresql.org Path: not-for-mail From: Vivek Khera Newsgroups: ml.postgres.performance Subject: Re: Hardware recommendations to scale to silly load Date: Tue, 02 Sep 2003 12:01:16 -0400 Organization: Khera Communications, Inc., Rockville, MD Lines: 32 Message-ID: References: <1061953146.18108.291.camel@haggis> <3F4E5DA9.6396.5C8916@localhost> <20030829140503.GA19388@libertyrms.info> NNTP-Posting-Host: yertle.kcilink.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: lorax.kciLink.com 1062518476 55279 216.194.193.105 (2 Sep 2003 16:01:16 GMT) X-Complaints-To: daemon@kciLink.com NNTP-Posting-Date: Tue, 2 Sep 2003 16:01:16 +0000 (UTC) User-Agent: Gnus/5.1002 (Gnus v5.10.2) XEmacs/21.4 (Portable Code, berkeley-unix) Cancel-Lock: sha1:TAOnzxFs7fZr5XIPt/lpX/VZU2w= X-Virus-Scanned: by amavisd-new at kciLink.com X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200309/4 X-Sequence-Number: 3370 >>>>> "AS" == Andrew Sullivan writes: AS> On Fri, Aug 29, 2003 at 12:05:03AM -0700, William Yu wrote: >> We should see a boost when we move to 64-bit Linux and hopefully another >> one when NUMA for Linux is production-stable. AS> According to the people who've worked with SGIs, NUMA actually seems AS> to make things worse. It has something to do with how the shared AS> memory is handled. You'll want to dig through the -general or AS> -hackers archives from somewhere between 9 and 14 months ago, IIRC. I knew my PhD research would one day be good for *something* ... The basic premise of NUMA is that you can isolate which data belongs to which processor and put that on memory pages that are local/closer to it. In practice, this is harder than it sounds as it requires very detailed knowledge of the application's data access patterns, and how memory is allocated by the OS and standard libraries. Often you end up with pages that have data that should be local to two different processors, and that data keeps being migrated (if your NUMA OS supports page migration) between the two processors or one of them just gets slow access. I can't imagine it benefiting postgres given its globally shared buffers. -- =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Vivek Khera, Ph.D. Khera Communications, Inc. Internet: khera@kciLink.com Rockville, MD +1-240-453-8497 AIM: vivekkhera Y!: vivek_khera http://www.khera.org/~vivek/