When I resume a previous run that didn't finish, is it possible to resume using a different number of MPI processes compared to when you started? Or will it cause problems?
Hi @qminor . The .resume files are parallellisation agnostic, so you should definitely be able to re-start a run with a different numbers of MPI processes.
When I resume a previous run that didn't finish, is it possible to resume using a different number of MPI processes compared to when you started? Or will it cause problems?