asyncio is Python's framework for concurrent I/O using async/await. It runs many I/O operations cooperatively on a single thread via an event loop — efficient for high-concurrency I/O-bound work (network calls, web servers) without the overhead of threads.
import asyncio
async def fetch_data(id): # an async function = a coroutine
print(f"start {id}")
await asyncio.sleep(1) # await a non-blocking I/O operation (yields control)
print(f"done {id}")
return id
asyncio.run(fetch_data(1)) # run a coroutine from sync code
An async def function is a coroutine; await suspends it at an I/O point, letting the event loop run other coroutines meanwhile — then resumes when the awaited operation completes.
# ❌ sequential — total time ≈ 3 seconds
async def slow():
await fetch_data(1) # wait 1s
await fetch_data(2) # then wait 1s
await fetch_data(3) # then wait 1s
# ✅ concurrent — total time ≈ 1 second (all run together)
async def fast():
await asyncio.gather(fetch_data(1), fetch_data(2), fetch_data(3))
asyncio.gather runs coroutines concurrently — while one awaits I/O, the others proceed. Three 1-second waits overlap into ~1 second total. This is the core value: overlapping many I/O waits on one thread.
Threading: OS-managed, preemptive, real threads (GIL-limited for CPU)
asyncio: single thread, COOPERATIVE — coroutines yield at `await` points
→ lighter weight (handle thousands of connections), no thread overhead,
but one CPU-heavy coroutine blocks everything (no preemption)
async def bad():
time.sleep(5) # ❌ BLOCKS the whole event loop (sync sleep)
result = heavy_compute() # ❌ CPU work freezes all coroutines
async def good():
await asyncio.sleep(5) # ✅ non-blocking
# offload CPU work to a thread/process pool
await loop.run_in_executor(None, heavy_compute)
Because it's cooperative and single-threaded, any blocking/CPU-heavy call freezes all coroutines. Use await-able async APIs, and offload blocking/CPU work to an executor.
✓ High-concurrency I/O: web servers (FastAPI), web scraping, many API/DB calls,
websockets, real-time apps — thousands of concurrent connections efficiently
✗ CPU-bound work — use multiprocessing instead (asyncio doesn't help)
asyncio is the modern approach to high-concurrency I/O in Python, powering frameworks like FastAPI and enabling efficient handling of thousands of simultaneous connections on a single thread — far lighter than thread-per-connection.
Understanding coroutines, await, concurrent execution with gather, and especially the cardinal rule (never block the event loop; offload CPU/blocking work) is essential for writing performant async Python.
Knowing when it fits (I/O-bound concurrency) versus when it doesn't (CPU-bound → multiprocessing) is key to using it effectively.