Xref: utzoo comp.unix.admin:2193 comp.periphs.scsi:2852
Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!sobeco!sdesmara
From: sdesmara@sobeco.com (s.desmarais)
Newsgroups: comp.unix.admin,comp.periphs.scsi
Subject: Re: Archive media choice DAT or MO?
Message-ID: <1991Jun12.000935.23902@sobeco.com>
Date: 12 Jun 91 00:09:35 GMT
References: <1991Jun6.034023.3359@newshost.anu.edu.au> <1991Jun9.212006.25244@en.ecn.purdue.edu> <1991Jun11.001118.10538@swdsrv.edvz.univie.ac.at>
Organization: Groupe Sobeco, Montreal, Canada
Lines: 23

In <1991Jun11.001118.10538@swdsrv.edvz.univie.ac.at> mah@wu-wien.ac.at (Michael Haberler) writes:
>In article <1991Jun9.212006.25244@en.ecn.purdue.edu>, ghg@en.ecn.purdue.edu (George Goble) writes:
>|> The Exabyte 8mm "8500" can move at near DAT speeds.  They also have an
>|> 8200-SX which has high speed positioning, but is more cumbersome than
>Yes, but with a really bizarre positionioning method. Do you want to
>write down the record number where your backup ended? I rather have a 
>file mark.

>- michael

Yup, good ol' file marks.  That's what we're using here.  We're putting file
marks on the exabyte tapes every 100M.  To restore from the end of a tape,
we just forward space file a couple of file, and start the restore from there.
Restoring from the end of an exabyte tape takes about 20 minutes.
We're keeping a list of files dumped so we know in which section of the
tape they are.

But I'd be interested in knowing more about that "record number" scheme.
-- 
Stephane M. Desmarais               sdesmara@sobeco.com  ou  sdesmara@sobeco.ca
Division STS - Groupe Sobeco Inc.   {uunet | mcgill-vision}!sobeco.com!sdesmara
505 boul Rene-Levesque Ouest        bur: (514) 878-9090 poste 297
Montreal, Quebec CANADA H2Z 1Y7     fax: (514) 875-2673  dom: (514) 733-3245