Beware that the figure of several million passes, quoted by
manufacturers, is misleading. An average single write-cycle of one tape
(i.e. a single time filling it from empty to full via migration and/or
reclamation) can involve hundreds of so-called "write-passes" - even if
it's only mounted once to migrate something huge, and is not subjected
to "shoeshining". Mount it many times to migrate small sets of data, and
then add a few restores, and those passes add up very quickly. Therefore
you can burn through your millions in a hurry, even with LTO4's "write
slowly" technology that greatly reduces shoeshining.
Worse, a tape that actually does shoeshine can wear out its millions of
passes in very short order. I suspect that nearly all tape-to-tape
reclamation shoeshines to some extent, as the output tape waits for TSM
to find the next file it wants to keep on the 50%-empty input tape.
Perhaps somebody smarter than me can figure out a way to measure
shoeshining, or how many "passes" a tape has had, but I don't think the
data exists. My suspicion is that if you could accurately count
"passes", we'd find that tapes actually last quite a bit longer than the
manufacturer's specs.
Due to drastic differences in the number of uses of a particular tape
over a period of time (say, a year), it's useless to use elapsed time as
a measure of how worn a tape is. We have very active and very slow
management classes, depending on the volatility of the actual data. The
tapes unlucky enough to be picked for a volatile type of data (email...)
get beat up pretty fast, while those lucky enough to go on long-term
vacation storing archive data will last forever. You also have to count
restores.
Also, I think one of the most important factors in predicting tape life
is how many times it is removed from the library and transported. This
subjects it to temperature and humidity changes, dust, electromagnetic
fields, mechanical shock, and who knows what other hazards of the
outside world. Tapes that stay locked inside a library in a
climate-controlled computer room will last considerably longer.
We track all errors via AIX erept and by mechanically scanning the TSM
activity log. I gave up on using status=private volumes, because the
long-term statistics it collects are of limited use, and have gone back
to plain scratch tapes which are much easier to manage. We do not count
errors against a tape volume which were obviously the drive's fault. A
tape is retired on its SECOND media error. Since we got our new library
with LTO4 technology in January 2008, we have had 0 (zero) tapes retired
for any reason, including media faults. Our only LTO4 tape faults of any
kind have been unload failures.
There's quite a bit of hocus-pocus here, about technology that is really
extremely reliable. Some of it is basically urban legend remembered from
9-track open-reel IBM 3420 days. The ultimate answer is of course 42.
http://en.wikipedia.org/wiki/Answer_to_Life,_the_Universe,_and_Everything
Roger Deschner University of Illinois at Chicago rogerd AT uic DOT edu
======I have not lost my mind -- it is backed up on tape somewhere.=====
>
>TSM does not keep information on a tape beyond its current use.
>Not sure about LTO, but they are similar to 3590 which we run to the
>first write error. In general this takes years, but I have no idea how
>many mounts.
>
>Andy Huebner
>-----Original Message-----
>From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
>M.Kiran Kumar
>Sent: Friday, August 08, 2008 12:54 AM
>To: ADSM-L AT VM.MARIST DOT EDU
>Subject: [ADSM-L] Reclamation Tape Count...
>
>Hi,
>
>Please let me know how to check the tape, how many times it is reclaimed
>in
>TSM. And how many times can we use a LTO Gen3 and LTO Gen4 Media??
>
>
>Regards,
>Kiran M
>
>
>
>
>
>
>
>Disclaimer:
>This email message (including attachments if any) may contain
>privileged, proprietary, confidential information, which may be exempt
>from any kind of disclosure whatsoever and is intended solely for the
>use of addressee (s). If you are not the intended recipient, kindly
>inform us by return e-mail and also kindly disregard the contents of the
>e-mail, delete the original message and destroy any copies thereof
>immediately. You are notified that any dissemination, distribution or
>copying of this communication is strictly prohibited unless approved by
>the sender.
>
>DQ Entertainment (DQE) has taken every reasonable precaution to minimize
>the risk of transmission of computer viruses with this e-mail; DQE is
>not liable for any damage you may sustain as a result of any virus in
>this e-mail. DQE shall not be liable for the views expressed in the
>e-mail. DQE reserves the right to monitor and review the content of all
>messages sent to or from this e-mail address
>
>
>This e-mail (including any attachments) is confidential and may be legally
>privileged. If you are not an intended recipient or an authorized
>representative of an intended recipient, you are prohibited from using,
>copying or distributing the information in this e-mail or its attachments. If
>you have received this e-mail in error, please notify the sender immediately
>by return e-mail and delete all copies of this message and any attachments.
>Thank you.
>
>
>Regional Health's mission is to provide and support health care excellence in
>partnership with the communities we serve.
>
>Note: The information contained in this message, including any attachments,
>may be privileged, confidential, or protected from disclosure under state or
>federal laws . If the reader of this message is not the intended recipient, or
>an employee or agent responsible for delivering this message to the intended
>recipient, you are hereby notified that any dissemination, distribution, or
>copying of this communication is strictly prohibited. If you have received
>this communication in error, please notify the Sender immediately by a "reply
>to sender only" message and destroy all electronic or paper copies of the
>communication, including any attachments.
>
|
