[gmx-users] Job Submission script for Multiple nodes GPUs

[Carsten Kutzner]

On 22 Aug 2014, at 18:09, Xingcheng Lin <linxingcheng50311 at gmail.com> wrote:

> There will be  an error if I did
> mpiexec -np 4 mdrun_mpi -s run.tpr -cpi state.cpt -cpo state.cpt -noappend
> -gpu_id 01 -nb gpu -ntomp 6
> 
> Incorrect launch configuration: mismatching number of PP MPI processes and
> GPUs per node.
> mdrun_mpi was started with 4 PP MPI processes per node, but you provided 2
> GPUs.
> 
> Somehow the system still considers it as a single node.
Then your MPI processes are not started correctly across the two
nodes.

> 
> Do you know how to solve it?
That depends on the MPI library you are using. Probably you need to specify
some kind of a hostfile  or machinefile with the names of all nodes.

Carsten

> 
> 
> 
>> Good afternoon,
>> 
>> I am trying to use multiple nodes to do GPU simulation, each node has two
>> GPUs and 12 CPUs mounted. Is there any submission script for doing that?
>> 
>> For single node I used:
>> 
>> mpiexec -np 2 mdrun_mpi -s run.tpr -cpi state.cpt -cpo state.cpt -noappend
>> -gpu_id 01 -nb gpu -ntomp 6
>> 
>> For 2 nodes I cannot use the script like
>> 
>> mpiexec -np 4 mdrun_mpi -s run.tpr -cpi state.cpt -cpo state.cpt -noappend
>> -gpu_id 0123 -nb gpu -ntomp 6
> The -gpu_id string refers to the GPU id?s _per node_, so you should also
> use -gpu_id 01 on two and more of these nodes.
> 
> Best,
>  Carsten
> -- 
> Gromacs Users mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.


--
Dr. Carsten Kutzner
Max Planck Institute for Biophysical Chemistry
Theoretical and Computational Biophysics
Am Fassberg 11, 37077 Goettingen, Germany
Tel. +49-551-2012313, Fax: +49-551-2012302
http://www.mpibpc.mpg.de/grubmueller/kutzner
http://www.mpibpc.mpg.de/grubmueller/sppexa