In the device class, you can provide an estimated capacity of the media.
This is not a limit, but rather is used to do calculations from. Until you
reach this threshold, the actual amount of data on a cartridge is specified
as a percentage of this amount. After you exceed this amount, the field is
used to show the actual amount of data written to the media. Percent
utilization at this time will show approximately 100% full (I have seen
99+%).
In all cases, ADSM is measured the amount of data that he physically wrote to
the media. If the data is uncompressed then it will be an actual byte count
of the files. If the device hardware compresses the data, this is done after
ADSM measures it, and so you can see some high numbers. On the other hand,
if you are doing compression on the client, (to minimize network impacts)
then the number ADSM is giving you is the compressed amount. This will make
the capacity of the cartridge look low. Since the amount of compression
varies from file to file, ADSM does not report back on the "true" number of
bytes (uncompressed) when you do client compression.
So.... where does this leave you? There are always debates about who has the
best compression algorithms.... I won't try to get into that here. I think
you really need to look at the issue from the perspective of "where is the
best place to compress the data given my environment." For most people, it
seems to be "I'll do it on the client, so that I can minimize the impact on
the network." (This of course is not true for everyone).
When you look at the media, I think that you could *Probably* generalize and
say that if I wrote the same set of files to two different tapes, and on one
set I compressed at the device, and on one set I compressed at the client,
that the result would be that I would take up the same amount of space on the
media. It's just that ADSM will report the number of bytes that were
physically written to the device, and so in the first case, the number will
be bigger than in the second.
Hope this helps. If you have any more questions, feel free to post them,
either to me or the list.
Jerry Lawson
jlawson AT thehartford DOT com
______________________________ Forward Header __________________________________
Subject: DDS2/DDS3 Capacity questions
Author: INTERNET.OWNERAD at SNADGATE
Date: 3/11/98 3:07 AM
Hi again,
maybe my question days ago (see end of mail) was confused. So I want to
ask it again in other words.
How can I find out the amount of data was written to a adsm tape after
the tape was marked full.
I think that the value from the ESTIMATED CAPACITY field is not true. I
used DAT Tapes with DDS2 and DDS2C format. After filling up both tapes
had nearly the same ESCAP value from aprox. 3.8 GB. I thought that the
value for DDS2C must br near 8 GB.
I got the same result when I used DDS3 and DDS3C (aprox. 12 GB)
Thank you for any help.
Stephan Rittmann
Karlsruhe, Germany
-------------------------------------------------------------------------
Hi all,
Hi all,
Im confused about the capacity handling for DAT DDS2 and DDS3 cartridges.
I defined the following deviceclasses:
DDS2 - for DDS2 format without compression
DDS2C - for DDS2 with hardwarecompression
DDS2 - for DDS3 format without compression
DDS3C - for DDS3 format with hardwarecompression
I filled a tape (devclas DDS3C) with uncompressed data. The data i used
was really uncompressed. For a test i zipped that data and reached a very
good compression rate. The END-OF-Volume was at aprox. 12GB. I reached
the same value when I used the DDS3 devclas.
The same happend when I use the DDS2C class. Only 3.6 GB on different
volumes.
Thanks for any ideas,
Stephan
===================END FORWARDED MESSAGE===================
|
