CODEWITHSUNDEEP
pythonmultiprocessing
alwbtc
alwbtc

Reputation: 29465

Multiprocessing of summing 2 lists doesn't work in python

The sum of sum1 and sum2 must be equal to 499500 in the end of this code, but it prints 0, why?

import multiprocessing

sum1 = 0
sum2 = 0
def list_append_1(out_list):
    global sum1
    for i in out_list:
        sum1 += i
    print "sum1: ", sum1

def list_append_2(out_list):
    global sum2
    for i in out_list:
        sum2 += i
    print "sum2: ", sum2


if __name__ == "__main__":
    lista_1 = [i for i in xrange(500)]# Number of random numbers to add
    lista_2 = [i for i in xrange(500,1000)]
    procs = 2   # Number of processes to create

    # Create a list of jobs and then iterate through
    # the number of processes appending each process to
    # the job list
    jobs = []
    process_1 = multiprocessing.Process(target=list_append_1, args=(lista_1,))
    process_2 = multiprocessing.Process(target=list_append_2, args=(lista_2,))
    jobs.append(process_1)
    jobs.append(process_2)

    # Start the processes (i.e. calculate the random number lists)
    for j in jobs:
        j.start()

    # Ensure all of the processes have finished
    for j in jobs:
        j.join()

    print sum1 + sum2

Upvotes: 0

Views: 349

Answers (1)

dano
dano

Reputation: 94951

This is happening because each of the tasks that you run using multiprocessing.Process end up with their own copy of the sum1 and sum2 variables, along with a third copy held by the parent process. This is one of the limitations of using processes instead of threads; memory isn't shared between the processes by default. You can get the sums back by using a multiprocessing.Queue to send the sums to the parent process:

import multiprocessing

def list_append(out_list, q):
    my_sum = sum(out_list)
    print "sum: ", my_sum
    q.put(my_sum)

if __name__ == "__main__":
    lista_1 = [i for i in xrange(500)]# Number of random numbers to add
    lista_2 = [i for i in xrange(500,1000)]
    procs = 2   # Number of processes to create

    # Create a list of jobs and then iterate through
    # the number of processes appending each process to
    # the job list
    jobs = []
    q = multiprocessing.Queue()
    process_1 = multiprocessing.Process(target=list_append, args=(lista_1, q))
    process_2 = multiprocessing.Process(target=list_append, args=(lista_2, q))
    jobs.append(process_1)
    jobs.append(process_2)

    # Start the processes (i.e. calculate the random number lists)
    for j in jobs:
        j.start()

    total = q.get() + q.get()

    # Ensure all of the processes have finished        
    for j in jobs:
        j.join()

    print total

Output:

sum1:  124750
sum2:  374750
499500

Upvotes: 6

Related Questions