The cluster workflow manager should employ clever handling of resource bottlenecks.

Typical resource bottlenecks and rules for clever handling of resource bottlenecks are:

1. staging of smaller datasets before larger datasets (log10-scale)
2. do not stage more than N datasets at once (for N = 2 or 3)

1. do not move more than N datasets at once (for N = 2 or 3)
2. use move operations for input/output style processings (on the fly processing during copy)

See: Cluster Queue Optimization