I take in 1.5 TB of client compressed data across 12 adsm servers nightly
my daily scheduled events are single tasks with a set period of time between
them (as you first did)
you can get fancy and write external scripts to watch things and start new
tasks once other tasks finish but...
I just assign set windows for each of the activities and that's the way it
goes
If I have a moment of idle time on an adsm server I view it as a great
success ;-)
the end users generally beat me to a pulp 24 hours a day pushing hot & cold
DB backups at me nightly and redo logs hourly all day long not to mention
the creation of alternate environments from archives of  existing ones

I set specific windows on my adsm servers for
client incremental processing (server scheduled client events)
migration of storage pools (disk to tape)
expire inventory
audit license
adsm DB backup
reclamation
storage pool backup (tape to tape)

If things get finished in their windows all is well, when they start to
exceed their windows I have to do something, either jump on users for doing
something massive and not giving me a heads up, look into client traffic,
look into the general health & well being of the adsm server... can also be
a good indicator of when to scale up the environment...

just a note on what I do, be it right or wrong, its just the way I deal with
it.

later,
       Dwight

> ----------
> From:         Dana Mitchell;[SMTP:Dana.Mitchell AT ING-DM DOT COM]
> Reply To:     ADSM: Dist Stor Manager
> Sent:         Tuesday, March 21, 2000 8:09 AM
> To:   ADSM-L AT VM.MARIST DOT EDU
> Subject:      Daily work scheduling
>
> Hello,
>
> I am running TSM 3.7.1 under OS/390 2.4 processing approx 120 gb daily of
> incremental backups, and am having trouble getting the daily storage pool
> tasks scheduled.   I need to be able to backup the primary disk pools to
> tape, run expiration, backup the DB, run reclaims, etc.
>
> Originally I had each command seperately scheduled, but this wasted time
> during the day, by having processes start at fixed times, regardless of
> the
> outcome of the previous process.  I am currently running with multiple
> commands, run by a script, with WAIT=YES coded for each, to have a process
> start as soon as it's predecessor finishes, but this approach is not very
> flexable, and difficult to monitor and debug.
>
> I am curious how others handle these tasks in a large environment.  How do
> you schedule them for the best device utilization throughout the day? How
> do you monitor them for successful and timely completion?  Any suggestions
> are welcome.
>
> Thanks in advance
> Dana Michell
> Equitable of Iowa
>