How to run different VASP tasks in parallel on a supercomputer node at the same time while running a USPEX task?

65 views
Skip to first unread message

zicheng zhao

unread,
Apr 13, 2023, 2:09:31 AM4/13/23
to USPEX
Dear all,

I want to know how to run different VASP tasks in parallel on a supercomputer node at the same time while running a USPEX task. The USPEX version is USPEX_2021.0.1. And what I set in input.uspex is:
stages : [vasp1 vasp2 vasp3 vasp4 vasp5]
numParallelCalcs : 10
numGenerations : 30
stopCrit: 6
}

#define vasp1
{type : vasp, commandExecutable : 'mpirun -np 10 vasp_std', kresol: 0.16}

#define vasp2
{type : vasp, commandExecutable : 'mpirun -np 10 vasp_std', kresol: 0.14}

#define vasp3
{type : vasp, commandExecutable : 'mpirun -np 10 vasp_std', kresol: 0.12}

#define vasp4
{type : vasp, commandExecutable : 'mpirun -np 10 vasp_std', kresol: 0.10}

#define vasp5
{type : vasp, commandExecutable : 'mpirun -np 10 vasp_std', kresol: 0.09}

I want to submit the five VASP tasks at the same time on a supercomputer. And below is my SBATCH file:
#SBATCH --cpus-per-task=10
#SBATCH --ntasks-per-node=4

echo Time is `date`
echo Directory is $PWD
echo This job runs on the following nodes:
echo $SLURM_JOB_NODELIST
echo This job has allocated $SLURM_JOB_CPUS_PER_NODE cpu cores.

cd $PWD

 uspex -r

However, I observed that the five VASP tasks can't be submitted simultaneously. And they run one by one.  And the error file in the CalcFold is:
 srun: step creation temporarily disabled, retrying(Requested nodes are busy).

Could you please tell me how to fix this problem? Thank you very much.

Best Regards,
Zicheng.

zicheng zhao

unread,
Apr 13, 2023, 10:33:50 AM4/13/23
to USPEX
Sorry, it is not 5, it should be 10. I want the 10 parallel VASP tasks submitted by USPEX to run simultaneously. Could you please tell  me how to achieve this?

#SBATCH --cpus-per-task=10
#SBATCH --ntasks-per-node=10


echo Time is `date`
echo Directory is $PWD
echo This job runs on the following nodes:
echo $SLURM_JOB_NODELIST
echo This job has allocated $SLURM_JOB_CPUS_PER_NODE cpu cores.

cd $PWD

 uspex -r
Reply all
Reply to author
Forward
0 new messages