Path: utzoo!attcan!uunet!dino!ceres!rlm
From: rlm@ceres.physics.uiowa.edu
Newsgroups: comp.sys.dec
Subject: 4mm Versus 8mm Tape Backup Subsystems
Message-ID: <21496.2524aa95@ceres.physics.uiowa.edu>
Date: 30 Sep 89 16:59:17 GMT
Organization: Department of Physics and Astronomy, University of Iowa
Lines: 37

I posted an inquiry a few weeks ago about 4mm DAT tape subsystems, but no one
seemed to have any experience. We have a VAX-3200 (Q-bus)+2x3100 (VMS) cluster 
with some 1.5 GB of disk which we need to backup with minimal operator 
intervention. The choices seem to be:

   o Buy an 8mm Exabyte system. These are available in several packages
     and are in wide use. I have been told that they are `fairly' reliable
     but are extremely slow if one wants to recover selected file(s) 
     located in the middle of the tape. Also, the reliability is apparently
     not terrific- I was told that it was comparable to a TK-50 (ouch!!).

   o Buy a 4mm DAT system. These are apparently just becoming available for
     VAX/VMS systems (although I haven't been able to contact anyone who has
     used one yet). They have only 60% capacity per tape compared with 8mm
     but are claimed to be more reliable (because of the digital recording 
     technique) and have much faster selective file recovery (20sec maximum).

We need to purchase something right away because at the moment we have NO 
backup device. Can anyone offer useful suggestions on this subject? I 
would be particularly interested in (a) First hand experience with a DAT 
subsystem in a DEC environment (VMS) and/or (b) A consumer guide to the 8mm
systems- which is considered the most reliable, or the best buy?

Thanks in advance


Robert Mutel
Department of Physics and Astronomy
University of Iowa
Iowa City, IA 52242

==============================================================================
Phone: (319) 335-1950         |  TWX: 910-525-1398  
SPAN: 7230::RLM (CERES)       |  Bitnet: rlm%iowa.Physics.UIowa.Edu@CUNYVM 
HEPnet: 47588::RLM (IOWA)     |  Internet: rlm@iowa.Physics.UIowa.Edu 
==============================================================================