Hi all out there,

we are running a ADSM/HSM client (2.1.0.6G) on a RS/6k R-24. Data from the
R-24 is backuped to an RS/6k running ADSM-Server 2.1.0.12/0.12

Especially on Monday mornings I notice a huge delay for dsmrecalls when
users need to access their data. The server administrator informed me,
there is also a "Storagepool migration" going onto the exactly the same
tape (3590 Tape) as the recall needs to read.

Hello IBM are you listening?

Is there any way to automagically stop storagepool migration, to let the
processes which need highest priority, run on that priority, because the
storagepool migration might take real long (more then one hour) and the
AIX-Client is the fileserver for a NEC SX-4 Supercomputer, which might
have ALL it's processors idleing, because all of the jobs are waiting for
the recalls. Canceling storagepool migration will fix the problem
immediately, but human action of the admin is needed ...

I could suggest, that storagepool migration stops migration to that
specific tape, perhaps goes to a different one, and recalls the needed
data from that tape.

I think satisfaction of my users is MORE important than easiness of
implementation of DSMSERV.

        Best regards

        Andreas

+----------------------------------------------------------------------+
|       Andreas Landhaeusser                                           |
|       Deutsche Forschungsanstalt fuer Luft- und Raumfahrt e.V. (DLR) |
|       Bunsenstr. 10                                                  |
|       D-37073 Goettingen                                             |
|       Germany                                                        |
|       Phone: +49 551 7092379  EMail: Andreas.Landhaeusser AT DLR DOT De     |
+----------------------------------------------------------------------+