py-mpi4py

py-mpi4py

This package provides Python bindings for the Message Passing Interface (MPI) standard. It is implemented on top of the MPI-1/MPI-2 specification and exposes an API which grounds on the standard MPI-2 C++ bindings.

Slurm Script

#!/bin/bash
#SBATCH --job-name=mpi4py-test # create a name for your job
#SBATCH -o mpi4py_job.o%j # Name of stdout output file (%j expands to jobId)
#SBATCH -p CUIQue # Queue name
#SBATCH --nodes=4 # node count (max 8 as of now)
#SBATCH --ntasks=16 # total number of tasks
#SBATCH --cpus-per-task=1 # cpu-cores per task
#SBATCH --time=00:10:00 # total run time limit (HH:MM:SS)


module load py-setuptools/58.2.0
module load py-mpi4py/3.1.2
module load python/3.8


mpirun python3.8 hello_mpi.py

hello_mpi.py Code

# hello_mpi.py:
# usage: python hello_mpi.py

from mpi4py import MPI
import sys

def print_hello(rank, size, name):
msg = "Hello World! I am process {0} of {1} on {2}.\n"
sys.stdout.write(msg.format(rank, size, name))

if __name__ == "__main__":
size = MPI.COMM_WORLD.Get_size()
rank = MPI.COMM_WORLD.Get_rank()
name = MPI.Get_processor_name()

print_hello(rank, size, name)

Submit the job to the scheduler as

sbatch pympiHelloScript.slurm

For managing your “job” , refer to this guide.

For more on parallel python, refer to their official documentation
https://mpi4py.readthedocs.io/en/stable/