Migration from disk storage pools is indeed performed by node, and
nodes with the largest filespace occupancy are the first to be migrated.

However, migration from sequential storage pools is not performed by
node, but rather by volume.  This is done to minimize mounts for
source volumes.  Sequential volumes selected for migration are those
which have been least recently referenced.

There is not a straightforward way to selectively transfer data
for a specific node from one sequential pool to another.  If you
know the volume(s) on which a particular node's data are stored,
you could use the MOVE DATA command to move files from selected
volumes to the new storage pool.

Dave Cannon
ADSM Development