ADSM-L

Re: Tape Performance

2001-09-27 07:43:27
Subject: Re: Tape Performance
From: Richard Sims <rbs AT BU DOT EDU>
Date: Thu, 27 Sep 2001 07:40:16 -0400
...
>The way I see it if you only have fast streaming performance, then sure your
>migrations and backups will be fast, but reclamation and file system recovery
>with the active versions of data spread across a tape will have poor
>performance.  Is this accurate?

Basically true; but anything that delays an operation (incremental backups,
migration process deferring to other system tasks, etc.) will make for
tape idle time and loss of streaming.

>Are there any reports that compare the various tape drives in 'real world'
>situations?

See the list archives, Tivoli web site whitepapers, Gartner analyses, etc.  I
index some of the comparisons in http://people.bu.edu/rbs/ADSM.QuickFacts :

  3590 vs. 9840 tape drives:  (see follow-on paper just below this one)
        http://www.storage.ibm.com/hardsoft/tape/3590/prod_data/3590perform.pdf
        (G522-2508)
  3590 vs. 9840, 3580 Ultrium, DLT 8000 tape drives:
        http://www.tivoli.com/news/press/analyst/tsm.pdf
  LTO Ultrium vs. Super-DLT:
        http://www.storage.ibm.com/hardsoft/tape/lto/prod_data/ltovsdlt.html

In general, tape technologies that operate without servo tracking suffer in
performance.

>And finally if there are no reports of 'real world' situations comparing the
>various tape drives...

See them prior postings in ADSM-L.  Lots of real experience and tears there.

   Richard Sims, BU
<Prev in Thread] Current Thread [Next in Thread>