Re: Running multi-node jobs on Sun Grid Engine

From: Éric Germaneau (germaneau_at_sjtu.edu.cn)
Date: Sun Apr 03 2016 - 20:34:20 CDT

Hey Sourav,

Here is how I run it with SLURM:

    NP=`srun --nodes=${SLURM_NNODES} bash -c 'hostname' |wc -l`
    MPIRUN=mpirun
    EXE=/path/to/namd/namd-2.10/namd2_CPU
    INPUT=apoa1.namd
    OUTPUT=namd.log
    $MPIRUN -np $NP $EXE $INPUT > $OUTPUT

By the way, did you successfully build the 2.11 release ?

On 04/03/2016 11:49 PM, Sourav Ray wrote:
> Greetings
>
> I have been trying to run multi-node jobs on Sun Grid Engine resource
> manager for quite some time but without any success, would really
> appreciate some assistance.
>
> Cluster details:
>
> 1. Operating system: Cent OS - 6.2
> 2. Cluster management utility - ROCKS cluster
> 3. Resource manager and scheduler - Sun Grid engine
>
> Attached: SGE single node script, PBS and some other scheduler's
> multi-node job submission script.
>
> Regards
> Sourav

-- 
Éric Germaneau (艾海克), Specialist
Center for High Performance Computing
Shanghai Jiao Tong University
Room 205 Network Center, 800 Dongchuan Road, Shanghai 200240 China
Email:germaneau_at_sjtu.edu.cn Mobi:+86-136-4161-6480 http://hpc.sjtu.edu.cn

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:17 CST