Python multiprocess profiling

barmaley picture barmaley · Jun 14, 2012 · Viewed 15.5k times · Source

I'm struggling to figure out how to profile a simple multiprocess python script

import multiprocessing
import cProfile
import time
def worker(num):
    time.sleep(3)
    print 'Worker:', num

if __name__ == '__main__':
    for i in range(5):
        p = multiprocessing.Process(target=worker, args=(i,))
        cProfile.run('p.start()', 'prof%d.prof' %i)

I'm starting 5 processes and therefore cProfile generates 5 different files. Inside of each I want to see that my method 'worker' takes approximately 3 seconds to run but instead I'm seeing only what's going on inside the 'start'method.

I would greatly appreciate if somebody could explain this to me.

Update: Working example based on accepted answer:

import multiprocessing
import cProfile
import time
def test(num):
    time.sleep(3)
    print 'Worker:', num

def worker(num):
    cProfile.runctx('test(num)', globals(), locals(), 'prof%d.prof' %num)


if __name__ == '__main__':
    for i in range(5):
        p = multiprocessing.Process(target=worker, args=(i,))
        p.start()

Answer

zigg picture zigg · Jun 14, 2012

You're profiling the process startup, which is why you're only seeing what happens in p.start() as you say—and p.start() returns once the subprocess is kicked off. You need to profile inside the worker method, which will get called in the subprocesses.