Bill writes:

> Now you can mark these readonly and send them offsite.  For the next
> week there will be 5,000 additional tapes to send offsite, and so on
> and so forth.  After a year there are 260,000 tapes offsite.  Now how
> does an administrator consolidate these tapes?

You have an excellent point and I can see how this can be impractical
for an environment like this.  The way the administrator would do some
consolidation of tapes is with "offsite" reclamation which as you know,
is handled in a different way than onsite reclamation.  Periodic
"offsite" reclamation will help consolidate this offsite data, but doing
it too often (threshold too low) can result in a lot of data movement
thrashing.

With your example of a node's data being spread over offsite collocated
copy storage pool volumes (let's say 20), the "offsite reclamation" of
these twenty tapes might only involve mounting a couple onsite primary
storage pool volumes (to copy from) and a couple new onsite copy storage
pool volumes (to write to).  The net result after the "offsite
reclamation" is 20 offsite tapes that have been consolidated onto two
onsite collocated tapes (plus probably still at least one offsite
tape) and only four tape mounts involved.  These offsite tapes can then
be reused after the REUSEDELAY time period has expired.

But yes, with your 5000 client, collocated copy storage pool example, it
still lends itself as being totally impractical because in addition to
5000 mounts for a given storage pool backup, you would add all the
mounts needed for periodic "offsite reclamations" not to mention all the
data movement going on.

Dave Crockett - ADSM development