CODEWITHSUNDEEP
pythonmultithreadingqueue
Sid
Sid

Reputation: 215

processing limited threads at a time in a thread list

I have a thread list in python code which is quiet big

           threadlist = [thread1, thread2, thread3..........threadn]

where n is greater than 200 and I need to process 10 threads at a time using python queue, how do I do that. Suggestions deeply appreciated.

Upvotes: 0

Views: 37

Answers (1)

Kevin
Kevin

Reputation: 76184

This seems like a great place to use the threading.Semaphore class. It's designed explicitly to allow only a limited number of threads to access a resource at one time. You create a Semaphore(10) in the main thread, then have each child thread call acquire() at the start of its execution, and release at the end. Then only ten will run at a time.

Here's an example which abstracts out the semaphore handling to a thread subclass; but you could just as easily do it yourself inside the target function, if you don't mind a little less encapsulation.

from threading import Thread, Semaphore
import time

class PatientThread(Thread):
    MAXIMUM_SIMULTANEOUS_THREADS = 10
    _semaphore = Semaphore(MAXIMUM_SIMULTANEOUS_THREADS)
    def run(self):
        PatientThread._semaphore.acquire()
        super().run()
        PatientThread._semaphore.release()

def exampleTargetFunc(x):
    print(f"Thread #{x} is starting.")
    time.sleep(1)
    print(f"Thread #{x} is completing.")

threads = []
for i in range(200):
    threads.append(PatientThread(target=exampleTargetFunc, args=(i,)))
for t in threads:
    t.start()
for t in threads:
    t.join()

Result:

Thread #0 is starting.
Thread #1 is starting.
Thread #2 is starting.
Thread #3 is starting.
Thread #4 is starting.
Thread #5 is starting.
Thread #6 is starting.
Thread #7 is starting.
Thread #8 is starting.
Thread #9 is starting.
<a one second pause occurs here...>
Thread #2 is completing.
Thread #0 is completing.
Thread #1 is completing.
Thread #10 is starting.
Thread #11 is starting.
Thread #12 is starting.
Thread #4 is completing.
Thread #5 is completing.
Thread #3 is completing.
Thread #7 is completing.
Thread #13 is starting.
Thread #6 is completing.
Thread #16 is starting.
Thread #14 is starting.
Thread #15 is starting.
Thread #17 is starting.
Thread #9 is completing.
Thread #8 is completing.
Thread #18 is starting.
Thread #19 is starting.
<... And so on for the next 20 seconds>

This demonstrates that threads 10, 11, and 12 did not start until 0, 1, and 2 finished. And likewise for threads 3-9 and threads 13-19.

Upvotes: 1

Related Questions