Date: Mon, 22 Jul 2002 12:04:08 +0200 From: Andreas Koch <koch@eis.cs.tu-bs.de> To: freebsd-stable@freebsd.org Subject: Re: 4.6-RC: Glacial speed of dump backups Message-ID: <20020722100408.GP26095@ultra4.eis.cs.tu-bs.de> In-Reply-To: <20020722081614.E367@gsmx07.alcatel.com.au> References: <20020606204948.GA4540@ultra4.eis.cs.tu-bs.de> <20020722081614.E367@gsmx07.alcatel.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
Peter, many thanks for the reply, I had already given up hope of seeing one at all :-) I have already been looking at the NetBSD dump shortly after my initial message. Their version has a _considerably_ improved internal buffering architecture. However, my time-limited porting efforts have not been successful: While I can get it to compile (after associating anonymous shared memory segments with temporary files), the dump process itself coredumps. When enabling the internal checks of the buffering system, numerous errors and failed asserts come up. So, it appears some more effort is required here. I plan to look into this more closely, but I am currently swamped at work. However, the potential gains seem to be worthwhile: On our preliminary benchmarks, NetBSD dump running on a low-end Athlon (from a 7.2K rpm SCSI disk) _easily_ beats the backup throughput of Solaris 2.8 dump (mirrored 15K rpm SCSI disks) on a U280R (2x USIII+ 1GHz). Andreas Koch -- for now making do with the dump | team kludge ... -- Andreas Koch Email : koch@eis.cs.tu-bs.de Technische Universit"at Braunschweig Phone : x49-531-391-2384 Abteilung Entwurf integrierter Schaltungen FAX : x49-531-391-5840 M"uhlenpfordtstr. 23, D-38106 Braunschweig, Germany * PGP key available * To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020722100408.GP26095>