Received: from localhost (maia-5.hub.org [200.46.204.182]) by postgresql.org (Postfix) with ESMTP id A97009FB872 for ; Mon, 11 Jun 2007 14:48:59 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.182]) (amavisd-maia, port 10024) with ESMTP id 75279-06 for ; Mon, 11 Jun 2007 14:48:51 -0300 (ADT) X-Greylist: delayed 00:26:42.640554 by SQLgrey-1.7.4 Received: from svr4.postgresql.org (svr4.postgresql.org [66.98.251.159]) by postgresql.org (Postfix) with ESMTP id CAE869FB981 for ; Mon, 11 Jun 2007 14:48:52 -0300 (ADT) X-Greylist: domain auto-whitelisted by SQLgrey-1.7.3 Received: from beaver.fastcrypt.com (beaver.fastcrypt.com [64.7.141.17]) by svr4.postgresql.org (Postfix) with ESMTP id 7EB8F5AF02A for ; Mon, 11 Jun 2007 17:22:09 +0000 (GMT) Received: from [192.168.1.14] (dcdsl.ebox.com [64.7.143.116]) by beaver.fastcrypt.com (Postfix) with ESMTP id 87FBD1580683; Mon, 11 Jun 2007 13:32:28 -0400 (EDT) In-Reply-To: <20070611153414.GK24327@phlogiston.dyndns.org> References: <47A154B5-496C-478A-B60D-4238AF12EAC6@fastcrypt.com> <20070611153414.GK24327@phlogiston.dyndns.org> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <330ACA8C-3FD0-45C1-BD56-DD68B8CC8343@fastcrypt.com> Cc: pgsql-performance@postgresql.org Content-Transfer-Encoding: 7bit From: Dave Cramer Subject: Re: How much ram is too much Date: Mon, 11 Jun 2007 13:22:04 -0400 To: Andrew Sullivan X-Mailer: Apple Mail (2.752.3) X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200706/267 X-Sequence-Number: 25300 Hi Andrew On 11-Jun-07, at 11:34 AM, Andrew Sullivan wrote: > On Mon, Jun 11, 2007 at 11:09:42AM -0400, Dave Cramer wrote: >>> and set them to anything remotely close to 128GB. >> >> Well, we'd give 25% of it to postgres, and the rest to the OS. > > Are you quite sure that PostgreSQL's management of the buffers is > efficient with such a large one? No, I'm not sure of this. > In the past, that wasn't the case > for relatively small buffers; with the replacement of single-pass > LRU, that has certainly changed, but I'd be surprised if anyone > tested a buffer as large as 32G. So does anyone have experience above 32G ? Dave > > A > > -- > Andrew Sullivan | ajs@crankycanuck.ca > The whole tendency of modern prose is away from concreteness. > --George Orwell > > ---------------------------(end of > broadcast)--------------------------- > TIP 7: You can help support the PostgreSQL project by donating at > > http://www.postgresql.org/about/donate