From owner-freebsd-hardware@FreeBSD.ORG Fri Mar 19 02:36:09 2004 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E44CF16A4CE for ; Fri, 19 Mar 2004 02:36:09 -0800 (PST) Received: from gaff.hhhr.ision.net (gaff.hhhr.ision.net [195.180.9.213]) by mx1.FreeBSD.org (Postfix) with ESMTP id 65CB443D2F for ; Fri, 19 Mar 2004 02:36:09 -0800 (PST) (envelope-from ohoyer@ohoyer.de) Received: from gaff.hhhr.ision.net (localhost [127.0.0.1]) by gaff.hhhr.ision.net (8.12.11/8.12.11) with ESMTP id i2JAaCdD045734; Fri, 19 Mar 2004 11:36:12 +0100 (CET) (envelope-from ohoyer@ohoyer.de) Received: from localhost (ohoyer@localhost)i2JAaADH045731; Fri, 19 Mar 2004 11:36:11 +0100 (CET) (envelope-from ohoyer@ohoyer.de) X-Authentication-Warning: gaff.hhhr.ision.net: ohoyer owned process doing -bs Date: Fri, 19 Mar 2004 11:36:10 +0100 (CET) From: Olaf Hoyer Sender: ohoyer@gaff.hhhr.ision.net To: Lanny Baron In-Reply-To: <1079673332.33813.79.camel@panda> Message-ID: <20040319113314.M45686@gaff.hhhr.ision.net> References: <20040318232348.BE86443D2D@mx1.FreeBSD.org> <405A6537.2070607@pacific.net.sg> <1079670664.33813.72.camel@panda> <405A7B25.8040306@pacific.net.sg> <1079673332.33813.79.camel@panda> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII cc: "freebsd-hardware@freebsd.org" Subject: Re: Multiprocessor system VS one processor system X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Mar 2004 10:36:10 -0000 On Fri, 19 Mar 2004, Lanny Baron wrote: > Hi Erich, > Yes you are right. A Server Board cannot be changed with the expectation > that the system to still run. > > But as I said, with real redundancy, as some of our customers do have, > such that if Server 1 died, Server 2 picks up immediately. The cost of > which, is substantially less than that of systems such as you imply. > Yes, loadbalancing for simple things like mail or http also increases the availability, so that you can pick simpler and therefore cheaper systems, when it comes to databases or things, that cannot be easily restored, some big machine should be preferred. But its always the question, how important the service/data is to the customer. In typical low-budget webhosting scenario, where a customer pays 1$/month or so, including a domain, the availability is not that important compared to travel agencies, money institutes etc... Olaf -- Olaf Hoyer ohoyer@gaff.hhhr.ision.net Fuerchterliche Erlebniss geben zu raten, ob der, welcher sie erlebt, nicht etwas Fuerchterliches ist. (Nietzsche, Jenseits von Gut und Boese)