2

[AndreiArgeanu]

2

[AndreiArgeanu]

After some research I have found the solution. It's a library called multiprocessing which seems to be included in python and doing some simple test I was able to see significant performance improvements using it and also got a noticeable increase of CPU usage from 30% to 85-97% on a 4C/8T i5 8250U. Using a perfr counter just running the calculation 4 times took 8.33 seconds, using multiprocessing with 4 processes I got 2.98 seconds which is a significant improvement.

 

Here's the Python docs if anyone is interested. It provides good examples and explanations. There's also YouTube videos but your mileage may vary.

https://docs.python.org/3/library/multiprocessing.html

 

This is the software I ran to test the performance

No multiprocessing

import time
start=time.perf_counter()

def do_something():
    x=10
    for i in range(22):
        x=x*x
    print("Done")

do_something()
do_something()
do_something()
do_something()

finish=time.perf_counter()

print(f"FInished in {round(finish-start, 2)} second(s)")

 

With multiprocessing

import time
import multiprocessing

start = time.perf_counter()

def do_something():
    x=10
    for i in range(22):
        x=x*x
    print("Done")


p1=multiprocessing.Process(target=do_something)
p2=multiprocessing.Process(target=do_something)
p3=multiprocessing.Process(target=do_something)
p4=multiprocessing.Process(target=do_something)

if __name__=='__main__':
    p1.start()
    p2.start()
    p3.start()
    p4.start()
    p1.join()
    p2.join()
    p3.join()
    p4.join()
    finish=time.perf_counter()

    print(f"FInished in {round(finish-start, 2)} second(s)")