All 15 Concurrency (asyncio/threading/multiprocessing) questions
More in Concurrency (asyncio/threading/multiprocessing)
What is the fundamental difference between concurrency and parallelism, and how does this map to Python's threading versus multiprocessing modules? How do you create and run a basic thread using the threading module? What is the basic pattern for running an async function using asyncio.run()? How do you run multiple coroutines concurrently using asyncio.gather()? How do you protect shared mutable state from race conditions when using multiple threads, using threading.Lock? How does multiprocessing.Pool let you parallelize a CPU-bound function across multiple worker processes? What is a race condition, and how does it differ from a deadlock? How does asyncio's event loop schedule and switch between multiple coroutines, and what makes this cooperative rather than preemptive? How do you safely run a blocking, synchronous function from within an asyncio coroutine without freezing the event loop? How do you share data between processes using multiprocessing, given that separate processes don't share memory the way threads do?
Advertisement
Python › Concurrency (asyncio/threading/multiprocessing)

How does the Global Interpreter Lock (GIL) affect CPU-bound versus I/O-bound multithreaded programs differently?

Intermediate

Answer

The GIL ensures only ONE thread executes Python bytecode at any given instant, so CPU-bound multithreaded code gets NO real speedup (and can even be slightly slower due to lock overhead); I/O-bound code, however, RELEASES the GIL during blocking operations (like network calls or file reads), so multiple threads CAN make real concurrent progress waiting on I/O.
# CPU-bound: threading gives little to no benefit due to the GIL
def cpu_work(): sum(i * i for i in range(10_000_000))

# I/O-bound: threading helps, since the GIL is released during network waits
def io_work(): requests.get("https://example.com")  # GIL released while waiting
Real-world example Choosing threading for a multi-URL web scraper (I/O-bound) but multiprocessing for a CPU-heavy data transformation job.

Common follow-ups: How does asyncio compare to threading specifically for I/O-bound concurrency, in terms of overhead?

The Global Interpreter Lock (GIL)
All 15 Concurrency (asyncio/threading/multiprocessing) questions