numba-mpi 0.40

Numba @njittable MPI wrappers tested on Linux, macOS and Windows

numba-mpi

Numba @njittable MPI wrappers

• covering: size/rank, send/recv, allreduce, bcast, scatter/gather & allgather, barrier and wtime
• basic asynchronous communication with isend/irecv (only for contiguous arrays); for request handling including wait/waitall/waitany and test/testall/testany
• not yet implemented: support for non-default communicators, ...
• API based on NumPy and supporting numeric and character datatypes
• auto-generated docstring-based API docs on the web: https://numba-mpi.github.io/numba-mpi
• pure-Python implementation with packages available at PyPI, Conda Forge and for Arch Linux
• CI-tested on: Linux (MPICH, OpenMPI & Intel MPI), macOS (MPICH & OpenMPI) and Windows (MS MPI)

Hello world example:

import numba, numba_mpi, numpy

@numba.njit()
def hello():
  print(numba_mpi.rank())
  print(numba_mpi.size())

  src = numpy.array([1., 2., 3., 4., 5.])
  dst_tst = numpy.empty_like(src)

  if numba_mpi.rank() == 0:
    numba_mpi.send(src, dest=1, tag=11)
  elif numba_mpi.rank() == 1:
    numba_mpi.recv(dst_tst, source=0, tag=11)

hello()

For information on MPI, see:

• https://www.mpi-forum.org/docs/
• https://mpi4py.readthedocs.io/en/stable/