2 methods running in Pool with same Input Values shows no multiprocessing

Question

I am trying to understand how 2 methods can be executed in parallel using a simple and dumbed down example. The issue is with respect to the execution results. I observe that the execution happens in sequential order where the method threeCheck() is executed in its entirety before sevenCheck() is even called. But my understanding was that both the methods would be executed in parallel.

threeCheck() is a method that checks if a number is divisible by 3
sevenCheck() is a method that checks if a number is divisible by 7
I have added a random sleep duration in the method calls to mimic latency.

I see that the inputs passed into threeCheck() Method are being processed in parallel. If we follow sequence, the numbers in the array divisible by 3 are : 84, 66, 60, 66, 87, 99, 42, ....... But when we observe the execution results, the sequence shows otherwise for these first results

import multiprocessing as mp
import time
import random

def threeCheck(inputParam):
    if inputParam % 3 == 0:
        duration = random.randint(0,6)
        print("3 INPUT = {0} and Sleep = {1}".format(inputParam, duration))
        time.sleep(duration)
        return inputParam

def sevenCheck(inputParam):
    if inputParam % 7 == 0:
        duration = random.randint(0,6)
        print("7 INPUT = {0} and Sleep = {1}".format(inputParam, duration))
        time.sleep(duration)
        return inputParam

def main():
    cpuCount = mp.cpu_count()
    pool = mp.Pool(cpuCount)
    inputValues = []
    for i in range(50):
        inputValues.append(random.randint(0,100))

    resultThree = pool.map(threeCheck, inputValues)
    resultSeven = pool.map(sevenCheck, inputValues)

    print("INPUT = ", inputValues)
    print("CPU = ", cpuCount)
    print("RESULT 3 = ", resultThree)
    print("RESULT 7 = ", resultSeven)

if __name__ == "__main__":
    main()

OUTPUT:
3 INPUT = 84 and Sleep = 5
3 INPUT = 66 and Sleep = 2
3 INPUT = 60 and Sleep = 1
3 INPUT = 99 and Sleep = 3
3 INPUT = 66 and Sleep = 3
3 INPUT = 42 and Sleep = 5
3 INPUT = 12 and Sleep = 1
3 INPUT = 87 and Sleep = 6
3 INPUT = 57 and Sleep = 5
3 INPUT = 42 and Sleep = 3
3 INPUT = 30 and Sleep = 2
3 INPUT = 63 and Sleep = 0
3 INPUT = 42 and Sleep = 6
3 INPUT = 24 and Sleep = 2
3 INPUT = 21 and Sleep = 6
3 INPUT = 30 and Sleep = 4
3 INPUT = 27 and Sleep = 0
7 INPUT = 7 and Sleep = 1
7 INPUT = 84 and Sleep = 6
7 INPUT = 42 and Sleep = 6
7 INPUT = 28 and Sleep = 1
7 INPUT = 21 and Sleep = 6
7 INPUT = 42 and Sleep = 5
7 INPUT = 63 and Sleep = 3
7 INPUT = 91 and Sleep = 5
7 INPUT = 42 and Sleep = 6
INPUT =  [7, 95, 25, 19, 59, 82, 47, 68, 20, 76, 84, 8, 88, 29, 1, 66, 60, 66, 87, 25, 23, 80, 52, 99, 65, 65, 25, 42, 12, 38, 28, 88, 57, 21, 27, 40, 42, 38, 31, 91, 30, 68, 22, 24, 63, 42, 64, 17, 65, 30]
CPU =  4
RESULT 3 =  [None, None, None, None, None, None, None, None, None, None, 84, None, None, None, None, 66, 60, 66, 87, None, None, None, None, 99, None, None, None, 42, 12, None, None, None, 57, 21, 27, None, 42, None, None, None, 30, None, None, 24, 63, 42, None, None, None, 30]
...

I am sure that I am missing something here ..

`Pool.map()` is blocking, so your understanding here was indeed wrong and the observed behaviour is what you should expect. Only the work _within_ the `.map()` call gets parallelized. If you need **two different** functions parallelized, you need to use [Pool.apply_async()](https://docs.python.org/3.7/library/multiprocessing.html#multiprocessing.pool.Pool.apply_async) instead ([example](https://stackoverflow.com/a/56076681/9059420)). — Darkonaut, Jul 18 '19 at 00:07
The other mystery, about unexpected ordering, is caused by chunking. You can read about this topic [here](https://stackoverflow.com/q/53751050/9059420). — Darkonaut, Jul 18 '19 at 00:08

2 methods running in Pool with same Input Values shows no multiprocessing

0 Answers0