From owner-freebsd-stable Mon Jul 22 20:14: 4 2002 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 125AE37B400 for ; Mon, 22 Jul 2002 20:14:02 -0700 (PDT) Received: from alcanet.com.au (mail3.alcanet.com.au [208.178.117.11]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7B15743E42 for ; Mon, 22 Jul 2002 20:14:00 -0700 (PDT) (envelope-from peter.jeremy@alcatel.com.au) Received: from gsmx07.alcatel.com.au (IDENT:root@localhost.localdomain [127.0.0.1]) by alcanet.com.au (8.12.4/8.12.4/Alcanet1.3) with ESMTP id g6N3DJ8R015086; Tue, 23 Jul 2002 13:13:19 +1000 Received: from gsmx07.alcatel.com.au (localhost [127.0.0.1]) by gsmx07.alcatel.com.au (8.12.3/8.12.3) with ESMTP id g6N3DJ2h039625; Tue, 23 Jul 2002 13:13:19 +1000 (EST) (envelope-from peter.jeremy@alcatel.com.au) Received: (from jeremyp@localhost) by gsmx07.alcatel.com.au (8.12.3/8.12.3/Submit) id g6N3DJ0C039624; Tue, 23 Jul 2002 13:13:19 +1000 (EST) Date: Tue, 23 Jul 2002 13:13:18 +1000 From: Peter Jeremy To: Matthew Dillon Cc: Andreas Koch , freebsd-stable@FreeBSD.ORG Subject: Re: 4.6-RC: Glacial speed of dump backups Message-ID: <20020723131318.F38313@gsmx07.alcatel.com.au> Mail-Followup-To: Matthew Dillon , Andreas Koch , freebsd-stable@FreeBSD.ORG References: <20020606204948.GA4540@ultra4.eis.cs.tu-bs.de> <20020722081614.E367@gsmx07.alcatel.com.au> <20020722100408.GP26095@ultra4.eis.cs.tu-bs.de> <200207221943.g6MJhIBX054785@apollo.backplane.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <200207221943.g6MJhIBX054785@apollo.backplane.com>; from dillon@apollo.backplane.com on Mon, Jul 22, 2002 at 12:43:18PM -0700 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On 2002-Jul-22 12:43:18 -0700, Matthew Dillon wrote: >Here are the preliminary results when I test this dumping /usr >to /dev/null: > > DUMP: finished in 140 seconds, throughput 6413 KBytes/sec (8 MB cache) > DUMP: finished in 144 seconds, throughput 6235 KBytes/sec (4 MB cache) > DUMP: finished in 234 seconds, throughput 3836 KBytes/sec (0 MB cache) Impressive. This is definitely much easier than trying to merge reads of adjacent blocks into one physical read to a scatter buffer. I'll do some experimenting when I have a chance. Two notes: 1) I thought I'd tried something similar during my initial investigations (VLB 486 with IDE disks). I found the cost of reading the extra data ate most of the savings from sensible block ordering. 2) At the time, someone (I didn't keep a record) commented that dump deliberately re-read inodes to try and minimise problems due to active filesystems. The first point probably isn't relevant any more. The second point means that your cache may make dumping active partitions more dangerous. (Since the cached data may not be relevant any longer). (Softupdate snapshots would help here, but they're not in -STABLE and I don't think Kirk's fixed a race-to-root deadlock yet). I suspect a cleaner (though much more effort) approach would be to make dump much chummier with the UFS code in the kernel so that it used the kernel FS buffer (with some hooks to prevent dump blowing the buffer cache for other processes). Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message