How does the GIL restriction affect the parallel work of Python programs?

GIL (Global Interpreter Lock) in CPython is a mechanism that ensures only one thread executes Python bytecode at a time. This affects parallel execution as follows:

• Multithreading does not improve performance for CPU-bound tasks. Even with multiple threads, they cannot utilize multiple CPU cores for Python code execution.
• I/O-bound tasks benefit from multithreading. Threads waiting for I/O release the GIL, allowing others to run.

To bypass GIL, often use:

• Multiprocessing (the multiprocessing module), where each process has its own interpreter and GIL.
• Use C extensions that can release the GIL during heavy computations.

Example:

import threading
import time

def cpu_bound():
    count = 0
    for _ in range(10**7):
        count += 1

threads = [threading.Thread(target=cpu_bound) for _ in range(4)]
start = time.time()
for t in threads:
    t.start()
for t in threads:
    t.join()
print(f"Time: {time.time() - start} seconds")

Despite 4 threads, execution time is almost the same as a single thread due to GIL.