From owner-freebsd-scsi Sat Jul 4 19:49:39 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA12087 for freebsd-scsi-outgoing; Sat, 4 Jul 1998 19:49:39 -0700 (PDT) (envelope-from owner-freebsd-scsi@FreeBSD.ORG) Received: from freebie.lemis.com (freebie.lemis.com [139.130.136.133] (may be forged)) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA12076 for ; Sat, 4 Jul 1998 19:49:30 -0700 (PDT) (envelope-from grog@freebie.lemis.com) Received: (from grog@localhost) by freebie.lemis.com (8.9.0/8.9.0) id MAA18320; Sun, 5 Jul 1998 12:18:30 +0930 (CST) Message-ID: <19980705121830.B358@freebie.lemis.com> Date: Sun, 5 Jul 1998 12:18:30 +0930 From: Greg Lehey To: Ustimenko Semen , David Kelly Cc: Harlan Stenn , dc-sage@dc-sage.org, scsi@FreeBSD.ORG, amanda-users@cs.umd.edu Subject: Re: DDS/2 tape specs References: <199807032016.PAA21647@nospam.hiwaay.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.91.1i In-Reply-To: ; from Ustimenko Semen on Sat, Jul 04, 1998 at 03:34:28PM +0700 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Saturday, 4 July 1998 at 15:34:28 +0700, Ustimenko Semen wrote: > On Fri, 3 Jul 1998, David Kelly wrote: >> Harlan Stenn writes: >>> My second test took 8191 `y' characters followed by a newline: >>> >>>> yes `cat 8ky` | dd bs=8k of=/dev/rst0 >>> >>> and I eventually killed after over 24 hours' elapsed time. >>> When I killed it, I think it claimed to have written nearly 10G to the tape. That's reasonable. You've chosen an almost ideal pattern for compression. >> 24 hrs for only 10G doesn't sound good. >> >> To a similar DDS-2 (with compression) drive I once used dd to copy >> /dev/null until something broke. Got 88G on a DDS-1 90m tape. Took >> about 8 hours. SGI Indy R5000, Irix 6.2, OEM SGI/Seagate/Archive DDS-2 >> tape drive. > > Yes, in good times it writes real fs data on 500-800 Kb/s. > But in worst, it goes on 20-80 Kb/s :-( > > Else total capacity and speed decrease after you use tape many > times without eraseing it and/or cleaning drive's head. No, the capacity stays the same. But if you don't clean the head, it'll have more and more soft errors, and the throughput will drop dramatically. > IMHO there are linear dependence between data speed and cartridge > capacity (becouse the tape speed is fixed). Tape speed is fixed, but that doesn't mean that it's moving all the time. If the drive gets a data underrun on write, it'll stop, back up, and start again. This can drop its speed by 95% or more. This will also happen on I/O errors. This is obviously what's been happening to you. > I'm useing SDT-7000 under 2.2.6-stable. And i can't say: > `It works perfectly!` :-( Not surprising. Greg -- See complete headers for address and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message