Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 5 Mar 2002 16:52:03 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 5 Mar 2002 16:51:54 -0500 Received: from www.fibrespeed.net ([216.168.105.33]:47604 "HELO mail.fibrespeed.net") by vger.kernel.org with SMTP id ; Tue, 5 Mar 2002 16:51:37 -0500 Date: Tue, 5 Mar 2002 16:51:33 -0500 From: "Michael T. Babcock" To: linux-kernel@vger.kernel.org Subject: Changing subsystem priorities Message-ID: <20020305215132.GG15698@godzilla.fibrespeed.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.24i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I was looking at the performance of one of my systems the other day and realised that it was often slowed by processes requesting disk I/O while other (often unrelated) processes were waiting to send network packets. For the sake of thinking out loud, would it be possible to allow priority changes on a subsystem basis within the kernel? That is to say, either tell the kernel that handling disk io is less important than handling network io (or vice versa) or perhaps that processes waiting for disk io be given their timeslices a little 'later' so that other non-disk-bound processes get a few more? I understand that this is entirely based on the priorities I have for one sserver but other people may wish to accomplish similar things with different priorities. Is this type of scheduler adjustment possible or even desirable? For what its worth, the system in question is running 2.2.20 until I find stable drivers for my DPT SmartRaid V controller for 2.4.x. -- Michael T. Babcock CTO, FibreSpeed Ltd. (Hosting, Security, Consultation, Database, etc) http://www.fibrespeed.net/~mbabcock/ - 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/