multiprocessing Module & Multi-Core CPU Parallelism

1Concept

`multiprocessing` bypasses the GIL entirely by spawning separate operating system processes, each with its own independent Python interpreter and private memory space. It scales CPU-bound computations across 100% of available CPU cores.

2Architecture Diagram

Master Process ---> Fork/Spawn Child Process 1 (Core 0, Private Memory)
               ---> Fork/Spawn Child Process 2 (Core 1, Private Memory)

3Code Example

import multiprocessing

def square_worker(n: int) -> int:
    return n * n

if __name__ == "__main__":
    cores = multiprocessing.cpu_count()
    print(f"Available CPU Cores: {cores}")
    data = [10, 20, 30, 40]
    results = [square_worker(x) for x in data]
    print(f"Processed Results: {results}")

4Expected Output

Available CPU Cores: 8
Processed Results: [100, 400, 900, 1600]

5Key Takeaways

• ✓Processes have separate memory spaces; mutations in one process are invisible to others.
• ✓Spawning processes incurs OS fork/spawn overhead; use process pools for batch jobs.
• ✓Always protect process entry points with `if __name__ == '__main__':` on Windows.