From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263974AbTDJAqf (for ); Wed, 9 Apr 2003 20:46:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S263973AbTDJAqf (for ); Wed, 9 Apr 2003 20:46:35 -0400 Received: from pop.gmx.net ([213.165.65.60]:20319 "HELO mail.gmx.net") by vger.kernel.org with SMTP id S263974AbTDJAqe (for ); Wed, 9 Apr 2003 20:46:34 -0400 From: "Oliver S." To: Subject: WG: questions regarding Journalling-FSes and w-cache reordering Date: Thu, 10 Apr 2003 02:58:15 +0200 Message-ID: <001001c2fefc$44ed47d0$0200000a@kimba> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.4510 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Importance: Normal Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > I'd like to build a reliable ~1 TB RAID 5 from IDE drives. I will > gladly give up performance for high reliability and low cost. Most ATA-RAIDs disable write-caching, and that's not really a performance -disadvantage with a decent drive-scheduler (afaik most OSes implement a cyclical-scan and not elevator-schedulers) when the HD has a linear mapping; and today's ATA-HDs all have linear mappings. Only with back-to-back writes can cause the HD to have a lap of honour.