Python Threading Locks and Semaphores — ELI5

Why threads need traffic rules — and what happens when two threads try to change the same thing at once.

Imagine you and your roommate share one bathroom. If you both try to walk in at the same time, chaos. So you have a rule: if the door is locked, you wait.

That locked door is a lock in programming.

When your Python program has multiple threads (like multiple workers doing things at the same time), they sometimes need to touch the same piece of data — like a shared bank balance. If Thread A reads the balance ($100), and Thread B also reads it ($100), and they both try to add $50, you might end up with $150 instead of $200. Both thought they were adding to $100.

A lock prevents this. Before touching the shared data, a thread “locks the door.” Any other thread that needs the data has to wait until the door is unlocked. One at a time. No confusion.

Now imagine the bathroom has three stalls. Up to three people can use it at once, but the fourth has to wait. That’s a semaphore — it’s a lock that allows a set number of threads through at the same time. Useful when you want some parallelism but not unlimited access.

The downside? If a thread forgets to unlock the door, everyone waits forever. And if Thread A is waiting for Thread B’s lock while Thread B is waiting for Thread A’s lock, nobody moves. That’s called a deadlock — the programming equivalent of two people blocking a hallway, each waiting for the other to step aside.

One thing to remember: Locks let only one thread in at a time, semaphores let a limited number in — both exist to stop threads from stepping on each other’s work.

pythonconcurrencythreading

See Also

  • Python Actor Model Why treating each piece of your program like a person with their own mailbox makes concurrency way less scary.
  • Python Aiocache Caching aiocache remembers expensive answers so your async Python app doesn't waste time asking the same question twice.
  • Python Aiofiles Async Io aiofiles lets your async Python program read and write files without freezing — because normal file operations secretly block everything.
  • Python Aiohttp Understand Aiohttp through an everyday analogy so Python behavior feels intuitive, not random.
  • Python Anyio Portability AnyIO lets your async Python code work with any async library — write once, run on asyncio or Trio without changes.

Related Topics

ab-testing How tech companies run thousands of experiments at once to improve their products — the scientific method applied to everything from button colors to recommendation algorithms. 3 levels → activation-functions Why neural networks need these tiny mathematical functions — and how ReLU's simplicity accidentally made deep learning possible. 3 levels → ai-agents-architecture How AI systems go from answering questions to actually doing things — the design patterns that turn language models into autonomous agents that browse, code, and plan. 3 levels →