Parallel versions of MD engines

So which are the parallel versions of simulation engines that we should use, for the different combinations of workflows & architectures?

Does each of the following executables then run in parallel in the respective HPC - if not please, do correct it and point out the executable that runs in parallel.

Amber:

On Archer -> pmemd.MPI
On Bluewaters -> sander
On Stampede -> /opt/apps/intel13/mvapich2_1_9/amber/12.0/bin/sander

Gromacs:

On Archer -> gmx mdrun
On Bluewaters -> gmx mdrun
On Stampede -> mdrun_mpi

In addition, what other action should be taken so as to have the simulation tasks run in parallel - is the "uses_mpi" : True the only step to be taken in this respect?

Thanks.

Comments (2)