Dear Amber Users,
I have maybe a rather silly question, but I couldn't find an answer
anywhere.
I'm "successfully" using PLUMED with pmemd.cuda from AMBER 20 and I'm very
happy with it,
but I can't find a way to instruct PLUMED to run in parallel (MPI) while
calling it from a serial pmemd.cuda run (within the mdin file, as usual).
I know it is possible to run parallel PLUMED within AMBER with the MPI
version of pmemd.cuda, but I'm pretty happy with the speed reached by the
serial AMBE run, I would only like to speed-up the PLUMED computations
unlocking its MPI capabilities (the version of PLUMED I compiled is
MPI-enabled of course). Switching to a cuda.MPI run will not solve the
problem, as the GPU is already far quicker than the PLUMED part...
As far as I understand this is possible in GROMACS, so I think it should be
possible also in AMBER, I just don't understand how to achieve the wanted
behaviour (serial GPU AMBER run + MPI PLUMED run).
Thank you all for you assistance,
best regards,
Domenico
*Domenico Marson*, Ph.D.
Assegnista di Ricerca | Postdoctoral Researcher
Università degli Studi di Trieste | University of Trieste
Department of Engineering and Architecture (DEA)
Molecular Biology and Nanotechnology Laboratory . UniTS (MolBNL.UniTs)
Via Valerio, 10 - 34127 Trieste (Italy)
domenico.marson.dia.units.it
+39 040 5583757
Skype: domenicomars
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Feb 12 2021 - 04:30:02 PST
