From owner-freebsd-stable Tue Mar 20 10:54:12 2001 Delivered-To: freebsd-stable@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by hub.freebsd.org (Postfix) with ESMTP id 8E50137B71A for ; Tue, 20 Mar 2001 10:54:07 -0800 (PST) (envelope-from bright@fw.wintelcom.net) Received: (from bright@localhost) by fw.wintelcom.net (8.10.0/8.10.0) id f2KIs5P24224; Tue, 20 Mar 2001 10:54:05 -0800 (PST) Date: Tue, 20 Mar 2001 10:54:04 -0800 From: Alfred Perlstein To: Gordon Tetlow Cc: Jan Conrad , freebsd-stable@FreeBSD.ORG Subject: Re: NFS performance Message-ID: <20010320105404.D29888@fw.wintelcom.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from gordont@bluemtn.net on Tue, Mar 20, 2001 at 10:45:54AM -0800 X-all-your-base: are belong to us. Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG * Gordon Tetlow [010320 10:47] wrote: > On Tue, 20 Mar 2001, Jan Conrad wrote: > > > We're making plans to upgrade our NFS server to FreeBSD-4.3 (including > > new disks...) and I would like to ask about the status of NFS v3? > > Why do you need NFSv3? Are there particular features in NFSv3 that you > need or are you just guessing that v2 > v3 and therefore better? > > > Currently, a standard NFS mount (4.3BETA) gives us a sequential writing > > speed of approx. 2Mb/s (just 400Mb dd'd data, IBM DTLA-307045 with > > softupdates) and reading speed of 9Mb/s (the maximum). > > What is the speed of your network? > > > Mounting via TCP gives us 2.5Mb/s. (Is this safe?) > > Why are you using TCP? If you are on a reliable LAN, use UDP. TCP should > be used for long haul NFS. There are lots of reasons for using UDP, if you > want me to go into them, I will. > > > Are there any other possibilities to increase the writing speed (by means > > of tuning NFS...)? > > Again, try UDP. Also, go pick up the O'Reilly book on NFS and NIS. Alot of > what I'm going to tell you is in there (indeed the book is on my desk at > work). Actually, from what I've been told, TCP allows for much larger requests than what UDP does, afaik UDP maxes out at 8k while tcp should be able to go to 32k (maybe 64k) and give possibly better performance. Plus each time you 'hickup' under a UDP mount it's a lot more painful because since it's possible for each RPC to be broken into several packets you have a lot more retransmition to do. However, as the manpages and sysadmin books say, NFS tuning is a black art and results under various setting may vary wildly. -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message