Blocking the Event Loop
Concurrency: lesson 14 of 15
One thread runs everything, so one rude call stops the world.
Lesson 14 of 15 · 5 min
Blocking the Event Loop
Step 1 of 9
threads1
An event loop is one thread. Whatever is running owns it until it gives it back.
The Idea
An event loop is cooperative: one thread, one task running at a time, and control only changes hands at an await. Concurrency comes from tasks that yield while they wait. A call that does not yield — a blocking read, a tight computation — holds the only thread, and everything else stops.
Real-World Example
One clerk serving a queue by taking each customer as far as their paperwork allows, then calling the next while forms are filled in. Concurrency, with one clerk. Then someone asks them to count a jar of coins.
The Code
import asyncio, time
async def gauge(depth):
await asyncio.sleep(0.1) # yields the thread to the loop
return depth
async def rude(depth):
time.sleep(0.1) # holds the only thread
return depth
async def race(worker):
start = time.perf_counter()
await asyncio.gather(worker(3), worker(7))
return round(time.perf_counter() - start, 1)
print(asyncio.run(race(gauge))) # 0.1 -- the two waits overlapped
print(asyncio.run(race(rude))) # 0.2 -- they queued upYour turn
Put the steps in the right order.
- The loop starts the second task while the first is waiting
- A task runs until it reaches an await
- The first task resumes where it left off
- Its awaited operation reports that it is ready
Mini quiz
1 / 3