Thread View: comp.arch.storage
7 messages
7 total messages
Started by Steven Bridge
Thu, 05 Oct 1995 15:53
exabyte compression questions
Author: Steven Bridge
Date: Thu, 05 Oct 1995 15:53
Date: Thu, 05 Oct 1995 15:53
45 lines
1951 bytes
1951 bytes
We are using a Lago datawheel with two exabyte drives.
After a drive failure, one of these has been replaced with
a drive supplied by Lago Systems described on the box as
a 'DRIVE ASM,8500C'.
I believe this performs hardware compression.
Certainly, if I write to it I cannot read it on the older drive.
Question (1) is : how can the hardware compression be turned off ?
It is connected to a scsi controller in an IBM RS/6000 running
AIX 3.2.5.
The drive is configured as 'Other SCSI Tape Drive'.
IBM engineers have suggested setting the DENSITY setting #1 to
15 or 21 via the software interface provided. Neither had the
required effect.
IBM documentation implies that IBM tape drives support a compression=no
attribute but this is not available for 'Other SCSI' tape drives.
I thought that the compression drive would be able to
read all tapes that had been written on the older drive.
But I find that of the 15 tapes I've tried, 3 can't be read.
We write a header file on each tape using :
dd of=/dev/rmt0.1 bs=1k con=sync
LABEL
Ctrl/d
The failed reads produce :
dd if=/dev/rmt0.1 bs=1k
dd: 0+0 records in.
dd: 0+0 records out.
or :
tctl -f /dev/rmt0.1 fsf 1
/dev/rmt0.1 fsf 1 The process failed.: There is an input or output error.
Question (2) is : does anyone have any idea how we can read these
tapes on an 8500C drive ? This is particularly important as we are
about to replace the 'standard' 8500 drive.
--
+----------------------------------------------------------------------+
Steven Bridge Systems Group, Information Systems Division
University College London
Re: exabyte compression questions
Author: Jonathan M. Elson
Date: Fri, 06 Oct 1995 03:07
Date: Fri, 06 Oct 1995 03:07
12 lines
1157 bytes
1157 bytes
Steven Bridge (ccaasub@ucl.ac.uk) wrote: : We are using a Lago datawheel with two exabyte drives. : After a drive failure, one of these has been replaced with : a drive supplied by Lago Systems described on the box as : a 'DRIVE ASM,8500C'. : I believe this performs hardware compression. Nope! it is the 8505 that has the extra compression processor added. The 8500 just has an empty slot where the compression board would go. You need to make them get you an 8505. The 8500 will get massive errors and garble when it tries to read a tape written on an 8505 when the compression was turned on.
Re: exabyte compression questions
Author: Lee Ward
Date: Fri, 06 Oct 1995 15:21
Date: Fri, 06 Oct 1995 15:21
23 lines
1613 bytes
1613 bytes
jmelson@artsci.wustl.edu (Jonathan M. Elson) wrote: >Steven Bridge (ccaasub@ucl.ac.uk) wrote: >: We are using a Lago datawheel with two exabyte drives. >: After a drive failure, one of these has been replaced with >: a drive supplied by Lago Systems described on the box as >: a 'DRIVE ASM,8500C'. >: I believe this performs hardware compression. > >Nope! it is the 8505 that has the extra compression processor added. >The 8500 just has an empty slot where the compression board would go. >You need to make them get you an 8505. The 8500 will get massive >errors and garble when it tries to read a tape written on an 8505 >when the compression was turned on. Exabyte does make an 8500C. It is indeed an 8500 with compression. The 8500C will read tapes written on 8200, 8500 and 8500C, at least. I've been told that 8505 and 8500C write/compress the same. Haven't verified it though. As for the original question of turning compression off, the drive is capable. How to do it on the IBM box I do not know. --Lee
Re: exabyte compression questions
Author: Richard Campbell
Date: Tue, 10 Oct 1995 13:45
Date: Tue, 10 Oct 1995 13:45
25 lines
1695 bytes
1695 bytes
In article <4526i6$b76@newsreader.wustl.edu>, jmelson@artsci.wustl.edu (Jonathan M. Elson) writes: |> Steven Bridge (ccaasub@ucl.ac.uk) wrote: |> : We are using a Lago datawheel with two exabyte drives. |> : After a drive failure, one of these has been replaced with |> : a drive supplied by Lago Systems described on the box as |> : a 'DRIVE ASM,8500C'. |> : I believe this performs hardware compression. |> |> Nope! it is the 8505 that has the extra compression processor added. |> The 8500 just has an empty slot where the compression board would go. |> You need to make them get you an 8505. The 8500 will get massive |> errors and garble when it tries to read a tape written on an 8505 |> when the compression was turned on. Wrong!, the 8500c was the predecessor to the 8505 and it DOES perform compression. However it's a full height device and I believe that the 8505 is a half height device. In either case I would want Lago to replace my drive with a like drive, you certainly don't want two different type drives in your Datawheel especially as the previous poster points out when a non compression drive tries to read a compressed tape. -- Rich Campbell, richc@ti.com
Re: exabyte compression questions
Author: Michael van Elst
Date: Wed, 11 Oct 1995 13:19
Date: Wed, 11 Oct 1995 13:19
17 lines
948 bytes
948 bytes
jmelson@artsci.wustl.edu (Jonathan M. Elson) writes:
>Steven Bridge (ccaasub@ucl.ac.uk) wrote:
>: a 'DRIVE ASM,8500C'.
^^^^^
>Nope! it is the 8505 that has the extra compression processor added.
>The 8500 just has an empty slot where the compression board would go.
Hmm. There is the EXB 8500C that does have the compression board.
Regards,
--
Michael van Elst
Internet: mlelstv@serpens.rhein.de
"A potential Snark may lurk in every tree."
Re: exabyte compression questions
Author: "John R. Vanderpool"
Date: Wed, 18 Oct 1995 21:37
Date: Wed, 18 Oct 1995 21:37
27 lines
1765 bytes
1765 bytes
richc@phantom.pac.sc.ti.com (Richard Campbell) writes:
>In article <4526i6$b76@newsreader.wustl.edu>, jmelson@artsci.wustl.edu (Jonathan M. Elson) writes:
>|> Steven Bridge (ccaasub@ucl.ac.uk) wrote:
>|> : We are using a Lago datawheel with two exabyte drives.
>|> : After a drive failure, one of these has been replaced with
>|> : a drive supplied by Lago Systems described on the box as
>|> : a 'DRIVE ASM,8500C'.
>|> : I believe this performs hardware compression.
>|>
>|> Nope! it is the 8505 that has the extra compression processor added.
>|> The 8500 just has an empty slot where the compression board would go.
>|> You need to make them get you an 8505. The 8500 will get massive
>|> errors and garble when it tries to read a tape written on an 8505
>|> when the compression was turned on.
> However it's a full height device and I believe that the 8505
>is a half height device.
that is correct.
fish
--
Life is what you make it, at least that's what the people say.
And if we can't make it through tomorrow, we'd better make it through today.
for Earth Science data and information use URL http://daac.gsfc.nasa.gov/
John R. Vanderpool <fish@daacdev1.stx.com> NASA/GSFC/HSTX
Re: exabyte compression questions
Author: "John R. Vanderpool"
Date: Wed, 18 Oct 1995 21:41
Date: Wed, 18 Oct 1995 21:41
30 lines
1924 bytes
1924 bytes
Lee Ward <lee> writes:
>jmelson@artsci.wustl.edu (Jonathan M. Elson) wrote:
>>Steven Bridge (ccaasub@ucl.ac.uk) wrote:
>>: We are using a Lago datawheel with two exabyte drives.
>>: After a drive failure, one of these has been replaced with
>>: a drive supplied by Lago Systems described on the box as
>>: a 'DRIVE ASM,8500C'.
>>: I believe this performs hardware compression.
>>
>>Nope! it is the 8505 that has the extra compression processor added.
>>The 8500 just has an empty slot where the compression board would go.
>>You need to make them get you an 8505. The 8500 will get massive
>>errors and garble when it tries to read a tape written on an 8505
>>when the compression was turned on.
>Exabyte does make an 8500C. It is indeed an 8500 with compression. The 8500C
>will read tapes written on 8200, 8500 and 8500C, at least. I've been told that
>8505 and 8500C write/compress the same. Haven't verified it though.
they use the same LZ and are fully compatible (i've tested it)
hi lee ;-) hey, your news From: line is bogus, or is that on purpose?...
fish
--
Life is what you make it, at least that's what the people say.
And if we can't make it through tomorrow, we'd better make it through today.
for Earth Science data and information use URL http://daac.gsfc.nasa.gov/
John R. Vanderpool <fish@daacdev1.stx.com> NASA/GSFC/HSTX
Thread Navigation
This is a paginated view of messages in the thread with full content displayed inline.
Messages are displayed in chronological order, with the original post highlighted in green.
Use pagination controls to navigate through all messages in large threads.
Back to All Threads