I would like to be able to prune the files that are kept on the ADSM
server in order to provide the kind of granularity of backup that we have
gotten with traditional backup schemes, without having to maintain
excessive numbers of copies of files.

The f0ollowing is an example of what I am after.

The goal is daily granularity for a week, weekly granularity for a month
and monthly for a year.

For each file backed up, I want to keep on the ADSM server:

-A daily copy of the file for the last  week (max 7 versions)
-One copy (the last) of the file for each of the last 4 weeks (max 4
-One copy (the last) of the file for each of the last 4 weeks (max 4
                copies)
                copies)

-One copy (the last) for each of the last 12 months (max 12 copies).
(Obviously, if the file hasn't been modified, the same file might act as
(Obviously, if the file hasn't been modified, the same file might act as
the "end-of-week" and "end-of-month" version for several months.)

I could do different schedules and capture end-of-week and end-of-month
copies, but the copies already exist on the ADSM server, so why spend the
time and network bandwidth to copy them again? (and again and again...)

I could also set my number of saved copies high enough to cover the
whole year, but if someone touched the file every day, then I would
potentially need 366 copies of every file, which seems excessive <gr>.

Too many copies is also a problem for restores, since it explodes the
number of files that must be reviewed to get the correct ones.

The simplest solution would be to prune the copies on the server on a
regular basis.  What I would like to do is run a job at the end of each
week that will delete all the excess copies.

Can I do this with ADSM?

Thanks for your help.

--
Mark Saltzman
Mark Saltzman
Assistant Director - Information Systems
University of Wisconsin-Extension
432 North Lake Street, Madison, Wisconsin  53706
TEL: 608.263.3084 / FAX: 608.262.2343