HP3000-L Archives

July 2001, Week 4

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Denys Beauchemin <[log in to unmask]>
Reply To:
Date:
Wed, 25 Jul 2001 16:05:24 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (72 lines)
Just a quick correction on what Steve said.  He is very correct about
everything, until he gets to the robotic libraries use of LUNs.  There are
indeed some robotic libraries that make use of LUNs.  These libraries are
very small, for example the SureStore 12000 (a 6 slot, DDS drive,) uses a
target ID for the entire device and then uses LUNs to differentiate between
the robot itself (what moves the tapes around,) and the data transfer
element (the tape drive.)  In this case, the SureStore uses LUN 0 for the
drive and LUN 1 for the robot.  This is one of a few devices that make use
of LUNs.  It does create problems.

The LUNs are never used to address slots, or disks, they are used to
address devices within the libraries.  One only needs to look at the SCSI
command set specification to see there is nothing that deal with slots and
CDs.

SCSI commands address the following devices:  Direct Access (disk drives,)
Sequential Access (tape drives,) Printer (lasers,) WORM, CD-ROM, Scanners,
Optical Memory, Media changer (robotic libraries,) and Communication
devices.  SCSI has nothing to do tapes, slots, platters, CD and paper.


Kind regards,

Denys. . .

Denys Beauchemin
HICOMP
(800) 323-8863  (281) 288-7438         Fax: (281) 355-6879
denys at hicomp.com                             www.hicomp.com


-----Original Message-----
From:   Steve Dirickson (Volt) [SMTP:[log in to unmask]]
Sent:   Wednesday, July 25, 2001 3:49 PM
To:     [log in to unmask]
Subject:        Re: Install New Drive - SCSI addresses...

> A clarification... I think that there is some confusion going
> on here about
> what the connection and configuration limits for SCSI are.  If my
> understanding is correct, the SE SCSI limit on the number of
> address values
> is NOT the same thing as the limit on the number of physical
> devices.  Again,
> if my understanding is correct, ONE SE SCSI address value can
> have up to 8
> subsidiary unit numbers (aka: "LUN" for Logical Unit Number)

Not quite. The point I've been trying to make is that the number of
possible target IDs HAS NOTHING TO DO with whether the bus is SE or not;
the number of target IDs is 8 for 8-bit SCSI and 16 for 16-bit SCSI.
Likewise, there's no such thing as "ONE SE SCSI address value"; a target
ID is a target ID; whether that target ID is on an SE bus, or a
differential bus, or an Ultra2 (LVD) bus, or an Ultra160 bus, or an
FC-AL bus, or any other kind of SCSI bus is irrelevant.

> Steve Dirickson wrote that the SE SCSI device limit is 56.

No, he didn't; I wrote that the 8-BIT BUS DEVICE LIMIT is 56, and that
the 16-BIT BUS LIMIT is 120.

FWIW, *very* few SCSI systems use multiple LUNs these days; I think some
robotic tape libraries use LUNs to identify the different tape slots,
and there are (or at least were) SCSI CD changers that used LUNs to
identify the different CD slots, but LUNs are a rarity.

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2