From owner-freebsd-current@FreeBSD.ORG Sun Jan 25 03:04:29 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 07ED116A4CE; Sun, 25 Jan 2004 03:04:29 -0800 (PST) Received: from happygiraffe.net (happygiraffe.net [81.6.215.59]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7F00E43D31; Sun, 25 Jan 2004 03:04:27 -0800 (PST) (envelope-from dom@happygiraffe.net) Received: from localhost (localhost.happygiraffe.net [127.0.0.1]) by happygiraffe.net (Postfix) with ESMTP id 02B4AB83E; Sun, 25 Jan 2004 11:04:26 +0000 (GMT) Received: from happygiraffe.net ([127.0.0.1]) by localhost (ppe.happygiraffe.net [127.0.0.1]) (amavisd-new, port 10024) with LMTP id 35528-04-2; Sun, 25 Jan 2004 11:04:25 +0000 (GMT) Received: by happygiraffe.net (Postfix, from userid 1001) id 52A31B80A; Sun, 25 Jan 2004 11:04:25 +0000 (GMT) Date: Sun, 25 Jan 2004 11:04:25 +0000 To: Robert Watson Message-ID: <20040125110425.GA35789@ppe.happygiraffe.net> References: <003c01c3de8d$d569edb0$471b3dd4@dual> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.5.1i From: dom@happygiraffe.net (Dominic Mitchell) X-Virus-Scanned: by amavisd-new at happygiraffe.net cc: performance@FreeBSD.ORG cc: current@FreeBSD.ORG Subject: Re: Old SUN NFS performance papers. X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 25 Jan 2004 11:04:29 -0000 On Sat, Jan 24, 2004 at 09:14:51PM -0500, Robert Watson wrote: > I haven't done much benchmarking on NFS lately, but something worth > remembering is that people have spent a lot of time researching and > optimizing TCP for a variety of connection types, whereas the NFS code > basically has a static implementation of RPC backoff and flow control that > hasn't evolved much. TCP is aware of things like the pathwise-mtu to the > server and adapts, whereas UDP just loses packets due to fragmentation, > especially if you are using larger block sizes. Please do post your > discoveries on performance@, and perhaps we could build an NFS performance > tuning section in the FreeBSD Handbook (or if there's not that much > content, add it to the FAQ)? I'm just playing with this... The first thing to note (probably) is to check that you can ping your server with a similiar size packet to the one you're using. I realised that my network isn't as robust as I thought it was very quickly yesterday, when pinging my server with an 8k packet. I was seeing 70% packet loss. The default ping showed no problems at all. The reason I mention it is that I'd been playing with NFS tuning because I had been seeing lockups. But the fault really lies at a lower level than NFS, it appears. -Dom