From owner-freebsd-questions@FreeBSD.ORG Thu Feb 26 13:33:15 2004 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4FCF516A4CE for ; Thu, 26 Feb 2004 13:33:15 -0800 (PST) Received: from smtpout.mac.com (A17-250-248-86.apple.com [17.250.248.86]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2B2BE43D2D for ; Thu, 26 Feb 2004 13:33:15 -0800 (PST) (envelope-from cswiger@mac.com) Received: from mac.com (smtpin08-en2 [10.13.10.153]) by smtpout.mac.com (Xserve/MantshX 2.0) with ESMTP id i1QLXEpS019194; Thu, 26 Feb 2004 13:33:14 -0800 (PST) Received: from [10.1.1.193] ([199.103.21.225]) (authenticated bits=0) by mac.com (Xserve/smtpin08/MantshX 3.0) with ESMTP id i1QLXDPP004257; Thu, 26 Feb 2004 13:33:13 -0800 (PST) In-Reply-To: <478667A6-6892-11D8-A5DD-00039367611E@obfuscated.net> References: <478667A6-6892-11D8-A5DD-00039367611E@obfuscated.net> Mime-Version: 1.0 (Apple Message framework v612) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <5FCEDFA8-68A3-11D8-870A-003065ABFD92@mac.com> Content-Transfer-Encoding: 7bit From: Charles Swiger Date: Thu, 26 Feb 2004 16:33:10 -0500 To: Michael Conlen X-Mailer: Apple Mail (2.612) cc: freebsd-questions@freebsd.org Subject: Re: NFS server usage X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 26 Feb 2004 21:33:15 -0000 On Feb 26, 2004, at 2:30 PM, Michael Conlen wrote: > Does FreeBSD's NFS implementation allow for caching of documents on > the client side, either its self or through the VM system's inactive > pages? Yes to both. NFS clients typically use something called biod or nfsoid, which implements some combination of caching and I/O request coalescing in order to reduce the amount of network traffic going to the server. [ Yes, this means that fsync() over NFS isn't guaranteed to actually have bits written to disk, at least historically... ] > The reason I'm asking is that I'm trying to size an NFS server using a > few of many similar sites that I hope to cluster. The performance so > far has been great, but I'm worried that there's something I'm missing > here that will cause the performance/usage to change in a very > nonlinear way. Any thoughts on the subject are appreciated. Well, you are going to be bottlenecked potentially by your network or by the maximum I/O rate that your NFS server can sustain. Your data suggests you ought to be able to handle about two orders of magnitude more net traffic, if you're over a dedicated 100 Mbs connection between server and clients (ie, using a switch), so it's likely that you're going to run into limits due to your disks well before then. You can probably switch to using rsync or some other replication scheme instead of NFS if you do run into limits, and keep the files locally if need be. -- -Chuck