I did the same thing, and you have to create a new storage pool in the new
library.  If you have a DASD pool to buffer the writes from VI on the
system, just point its NEXTPOOL parameter to the new storage pool.  If you
don't have a DASD pool, then point the NEXTPOOL from the existing library
to the pool in the new library.  When the first pool fills to the high
migration threshold, it will push data to the second pool.

ADSM/TSM handles the tracking of where the VI object is stored, so this is
all seamless to the users.

Nick Cassimatis
nickpc AT us.ibm DOT com

Today is the tomorrow of yesterday.