From owner-freebsd-questions@FreeBSD.ORG Wed Jul 23 22:38:07 2003 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 E2EA037B410 for ; Wed, 23 Jul 2003 22:38:07 -0700 (PDT) Received: from mail.seekingfire.com (coyote.seekingfire.com [24.72.10.212]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1BC8643FA3 for ; Wed, 23 Jul 2003 22:38:07 -0700 (PDT) (envelope-from tillman@seekingfire.com) Received: from blues.seekingfire.prv (blues.seekingfire.prv [192.168.23.211]) by mail.seekingfire.com (Postfix) with ESMTP id 0DB648F for ; Wed, 23 Jul 2003 23:38:06 -0600 (CST) Received: (from tillman@localhost) by blues.seekingfire.prv (8.11.6/8.11.6) id h6O5c5M24482 for freebsd-questions@freebsd.org; Wed, 23 Jul 2003 23:38:05 -0600 Date: Wed, 23 Jul 2003 23:38:05 -0600 From: Tillman To: FreeBSD-Questions Message-ID: <20030723233805.H20356@seekingfire.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i X-Urban-Legend: There is lots of hidden information in headers Subject: ng_one2many and switches 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, 24 Jul 2003 05:38:08 -0000 Howdy, I'd like to confirm my understanding of how ng_one2many works: Using a setup like http://bsdvault.net/sections.php?op=printpage&artid=98), I think that: * with an unmanaged switch I'll be able to effectively double the outgoing bandwidth on a given server, though my incoming bandwidth will remain at 100Mbit * with a managed switch that supports link aggregation I'll be able to effectively double the bandwidth in both directions * on a hub, of course, there would be no bandwidth increase i.e., this will be a good way to increase the performance of an NFS file server to multiple simultaneous clients on an network with unmanaged switches. The NFS server is currently network I/O limited - the vinum array runs almost four times as fast as I can serve the files, according to bonnie++ benchmarking runs. The NFS traffic is fairly async, being mostly reads. Does that sound about right? - Tillman -- "bash awk grep perl sed df du, du-du du-du, vi troff su fsck rm * halt LART LART LART!" - The Swedish BOFH