Book Image

Python Parallel Programming Cookbook

By : Giancarlo Zaccone
Book Image

Python Parallel Programming Cookbook

By: Giancarlo Zaccone

Overview of this book

This book will teach you parallel programming techniques using examples in Python and will help you explore the many ways in which you can write code that allows more than one process to happen at once. Starting with introducing you to the world of parallel computing, it moves on to cover the fundamentals in Python. This is followed by exploring the thread-based parallelism model using the Python threading module by synchronizing threads and using locks, mutex, semaphores queues, GIL, and the thread pool. Next you will be taught about process-based parallelism where you will synchronize processes using message passing along with learning about the performance of MPI Python Modules. You will then go on to learn the asynchronous parallel programming model using the Python asyncio module along with handling exceptions. Moving on, you will discover distributed computing with Python, and learn how to install a broker, use Celery Python Module, and create a worker. You will understand anche Pycsp, the Scoop framework, and disk modules in Python. Further on, you will learnGPU programming withPython using the PyCUDA module along with evaluating performance limitations.
Table of Contents (13 chapters)
Python Parallel Programming Cookbook
About the Author
About the Reviewers

Collective communication using Alltoall

The Alltoall collective communication combines the scatter and gather functionality. In mpi4py, there are three types of Alltoall collective communication:

  • comm .Alltoall(sendbuf, recvbuf): The all-to-all scatter/gather sends data from all-to-all processes in a group

  • comm.Alltoallv(sendbuf, recvbuf): The all-to-all scatter/gather vector sends data from all-to-all processes in a group, providing different amount of data and displacements

  • comm.Alltoallw(sendbuf, recvbuf): Generalized all-to-all communication allows different counts, displacements, and datatypes for each partner

How to do it…

In the following example, we'll see a mpi4py implementation of comm.Alltoall. We consider a communicator group of processes, where each process sends and receives an array of numerical data from the other processes defined in the group:

from mpi4py import MPI
import numpy

size = comm.Get_size()
rank = comm.Get_rank()

a_size = 1
senddata = (rank...