Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263646AbUAZPQO (ORCPT ); Mon, 26 Jan 2004 10:16:14 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263679AbUAZPQO (ORCPT ); Mon, 26 Jan 2004 10:16:14 -0500 Received: from opersys.com ([64.40.108.71]:24325 "EHLO www.opersys.com") by vger.kernel.org with ESMTP id S263646AbUAZPQM (ORCPT ); Mon, 26 Jan 2004 10:16:12 -0500 Message-ID: <4015303F.1030003@opersys.com> Date: Mon, 26 Jan 2004 10:20:31 -0500 From: Karim Yaghmour Reply-To: karim@opersys.com Organization: Opersys inc. User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030624 Netscape/7.1 X-Accept-Language: en-us, en, fr, fr-be, fr-ca, fr-fr MIME-Version: 1.0 To: Rik van Riel CC: Nuno Silva , JustFillBug , linux-kernel@vger.kernel.org Subject: Re: [ANNOUNCE] Cooperative Linux References: In-Reply-To: Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1868 Lines: 48 Rik van Riel wrote: >>So, for example, Xen assumes that all OSes are going to use the same >>devices for I/O: same disk, same NIC, etc. It therefore implements lots >>of virtual devices for these. > > > Consolidation means more efficient hardware use ... In the case of a UP system, it may ... depending on what you're trying to do. On an SMP system or an SMP-cluster, however, consilidation is likely to mean loss of performance. >>Wouldn't it be just better to reuse the existing work on the hotplug >>hardware (hotplug CPU, hotplug memory, etc.) to have the kernels >>get/return hardware resources to the nanokernel? > > > That means a loss of flexibility. It depends on your setup. In the case of an SMP-cluster where all OS instances are launched at startup, where runtime setup modification can be costly because of global table changes, and where centralization is to be avoided in as much as possible, then the hotplug capabilities are probably the best way to go. If there's a desire to have both capabilities (fine-grain allocation and gross-grain allocation) in whatever is finally adopted, then that's something to keep in mind. I guess I'm just saying that there are pros and cons, depending on your setup. > Furthermore, these hotplug > patches don't seem ready yet. Yes, I'm aware of this. There are other components of a truely generic virtualization interface which are missing. Does that mean we shouldn't think ahead? Karim -- Author, Speaker, Developer, Consultant Pushing Embedded and Real-Time Linux Systems Beyond the Limits http://www.opersys.com || karim@opersys.com || 1-866-677-4546 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/