Python Parallel Programming Cookbook

How to manage a state between processes

Python multiprocessing provides a manager to coordinate shared information between all its users. A manager object controls a server process that holds Python objects and allows other processes to manipulate them.

A manager has the following properties:

• It controls the server process that manages a shared object
• It makes sure the shared object gets updated in all processes when anyone modifies it

How to do it...

Let's see an example of how to share a state between processes:

1. First, the program creates a manager list, shares it between n number of taskWorkers, and every worker updates an index.
2. After all workers finish, the new list is printed to stdout:

   import multiprocessing def worker(dictionary, key, item): dictionary[key] ...