Multicomputer scheduling is something similar to the multiprocessor scheduling.
The simplest multiprocessor algorithm, that is, maintaining a single central list of ready processes, doesn't work however, since each and every process can only run on the central processing unit it is currently located on.
But however, whenever a new process is created, then a choice can be made that where to place the process.
Given that each node has their own processes, any local scheduling algorithm can be used.
However, gang scheduling can also be used here just in the same way as used in case of multiprocessor since that merely requires an initial agreement on which process to run in which time slot, and some way to coordinate the start of the time slots.