[gmx-users] Something is wrong with parallel runs

Anton Feenstra feenstra at few.vu.nl
Tue May 3 19:52:57 CEST 2005


Krewson, Tom wrote:

> I don't know what is wrong but when I try to do a parallel run using the
> mdrun_mpi program and give the np value to say 9, the output of gromacs
> always says: NNODES=1, MYRANK=0. Lamboot has booted all nodes with bhost
> file successfully. The simulation runs just fine but only on one node.
> Mdrun_mpi is on all nodes as well as mpi fftw and when I ssh to a
> compute node I can run lamboot and mdrun_mpi without a problem. Do the
> non mpi fftw and gromacs binaries need to be on each compute node for
> the parallel run to work? I must be missing something easy.

Did you:
- specify '-np 9' also for grompp?
- try something like 'mpi_run -np 9 mdrun_mpi -np 9'?


-- 
Groetjes,

Anton

* NOTE: New Phone & Fax numbers (below) *

  _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 59 87608 - Fax: +31 20 59 87610           |
|             | Feenstra at few.vu.nl - www.few.vu.nl/~feenstra/         |
|             | "If You See Me Getting High, Knock Me Down" (RHCP)    |
|_____________|_______________________________________________________|





More information about the gromacs.org_gmx-users mailing list