Python Threading/ThreadPool implementation

Question

I have the following two snippets showing the power of threading and was wondering what the difference is for each implementation.

from multiprocessing.dummy import Pool as ThreadPool

def threadInfiniteLoop(passedNumber):
    while 1:
        print passedNumber

if __name__ == '__main__':
    packedVals={
        'number':[0,1,2,3,4,5,6,7,8,9]
    }
    pool = ThreadPool(len(packedVals['number']))
    pool.map(func=threadInfiniteLoop,iterable=packedVals['number'])

and

import threading

def threadLoop(numberPassed):
    while 1:
        print numberPassed

if __name__ == '__main__':
    for number in range(10):
        t = threading.Thread(target=threadLoop, args=(number,))
        t.start()

What is the difference between the two snippets and their initialization's of each thread? Is there a benefit of one over the other and what would be a desirable situation where one would be more applicable than the other?

Answer 1

There's not much difference when you want to start a thread that runs forever.

Normally, you use a thread pool when your program continually creates new finite tasks to perform "in the background" (whatever that means).

Creating and destroying threads is relatively expensive, so it makes more sense to have a small number of threads that stick around for a long time, and then use those threads over and over again to perform the background tasks. That's what a thread pool does for you.

There's usually no point in creating a thread pool when all you want is a single thread that never terminates.