spawn n workers with GNU parallel without input

Question

I would like to trigger n jobs doing the following work:

# work.py
import time
import random
while True:
    print(random.randint(0,9))
    time.sleep(1)

where n is the number of available cores.

How can GNU parallel be used in such situation?

Answer 1

You should be able to use:

seq $(parallel --number-of-cores) |
    parallel -N0 --line-buffered python3 -u ./work.py

But I wonder why you don't use Python's built-in multiprocessing, rather than introduce an additional dependency.