The worst thing about your test is that it is not real world for a TSM
environment.  You won't be doing "full" backups so the test is not
representative.  Write performance to tape is determined by many factors.
In TSM, cataloging is done at the file level.  In most cases, the catalog
operation is taking place at intervals of 256 files (the max as specified by
txngroupmax).  So if your test disk has lots of little files TSM will spend
more time than Legato cataloging information.  This is a good thing for
restores, a place where TSM excels, and a bad thing for tape throughput (in
some cases).  If you want to see TSM move bytes to tape as fast as Legato,
get a single very large file.  Back it up with both tools.  I would guess
you will see very similar data transfer rates.

Also to consider: it's a real good thing that Legato works the way it does
since it needs to perform "full" backups regularly in order to offer
reasonable restores.  If it couldn't do fulls pretty fast you wouldn't be
able to get your work done.  TSM backs up less data on average (over time),
it catalogs more efficiently and it will outperform Legato on your most
typical restores.

Did I mention that the problem isn't backup, but rather restore?  If you
really want to evaluate the products, do some restore tests.  And don't
choose a test that you won't be doing very often in real life.  Choose a
test you will be doing in real life everyday: a restore of a single file.
By them time you and Legato have figured out which tape the data is on (and
retrieved it from offsite), TSM will have restored the file and fixed your
coffee.

Kelly J. Lipp
Storage Solutions Specialists, Inc.
PO Box 51313
Colorado Springs CO 80919
(719)531-5926
Fax: (719)260-5991
www.storsol.com
lipp AT storsol DOT com