Skip to main content

Run many jobs using MPI processes through a master/slave approach

Project description

A Python3 library for running many jobs on multiple processes, using MPI.

Example

The following example runs 3 jobs. For each of them, a message is printed on the screen.

from mpi4py import MPI
from time import sleep

import mpijobs

class MyJob(mpijobs.Job):
    def __init__(self,
                 params):
        super().__init__()
        self.params = params
        self.result_str = None

    def run(self, rank):
        print("Hi, I am {0} (process {1})".format(self.params, rank))

        # Pretend you're doing something useful
        sleep(1.0)

        # Perform a simple computation and save the result
        self.result_str = self.params.upper()

JOB_LIST = [MyJob(name) for name in ('Bob', 'Alice', 'Stephen')]
RESULTS = mpijobs.run_event_loop(MPI.COMM_WORLD, JOB_LIST)

if MPI.COMM_WORLD.rank == 0: # Only one process should print the results
    print("strings computed by the processes = [{0}]"
          .format(', '.join([x.result_str for x in RESULTS])))
    print("elapsed_time = [{0}]"
          .format(', '.join([str(x.running_time) for x in RESULTS])))

The program must be run using mpirun (or an equivalent program, depending on your MPI library):

mpirun -n 3 python3 example.py

The output is the following (actual ranks and times might vary):

Hi, I am Stephen (process 2)
Hi, I am Alice (process 1)
Hi, I am Bob (process 1)
strings computed by the processes = [ALICE, STEPHEN, BOB]
elapsed_time = [0:00:01.001056, 0:00:01.001088, 0:00:01.001041]

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

mpijobs-0.2.2-py3-none-any.whl (6.8 kB view details)

Uploaded Python 3

File details

Details for the file mpijobs-0.2.2-py3-none-any.whl.

File metadata

File hashes

Hashes for mpijobs-0.2.2-py3-none-any.whl
Algorithm Hash digest
SHA256 e33d5f56c51297d185eb07bfaad473e44278a4bc4fc9d0fdf4240a5b69c59ada
MD5 61d0b2de1d5e73f86bd051405be9cd34
BLAKE2b-256 eca9f3dd8a990063a6ed70329688f4af78017c66cb6d3dc72e960ba8052e2e90

See more details on using hashes here.

Provenance

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page