Concurrent Futures

A more modern and flexible approach can be found in the standard libraries concurrent.futures module.

The usage is quite similar to the way we used multiprocessing before.

Now we’re chosing an appropriate Executor to do the job.

Import section:

# Instead
import multiprocessing as mp

# do
import concurrent.futures as cf

Main block:

# Instead
with mp.Pool() as pool:
    iterations = pool.map(my_mandel_row, y)

# do
with cf.ProcessPoolExecutor() as executor:
    iterations = executor.map(my_mandel_row, y)

If you’re adventoureous use

with cf.ThreadPoolExecutor() as executor:
    iterations = executor.map(my_mandel_row, y)

to try your luck with a Threads based exeutor and enjoy the GIL