-Reply

I really like the idea of some way for duplicate file detection.  In a
typical workstation client environment, preloaded with apps
(SUN,Windows, and NT for us), more than 90% of the data may be
duplicated.  This is kinda depressing--thinking that 45 of 50 GB of data
in a storage pool is redundant, for instance.

In general we define base configurations for each of our client
environments and duplicate them dozens or hundreds of times
over--usually at a departmental boundary.  I have often thought that an
ADSM configuration option of "incremental based on <node>" would work
well for our environment.  In this case <node> would be a backup of the
virgin environment to be deployed.  For example, an ADSM options file
for members of the HR department might read "INCRemental based on
BASE_HRD".

Dennis Taylor
Georgetown University Hospital
taylordc AT gunet.georgetown DOT edu