Skip to content

Module 05 / Project 02 — Concurrent Requests

Home: README · Module: Async Python

Learn Your Way

Read Build Watch Test Review Visualize Try
— This project — — Flashcards — —

Focus

  • aiohttp for async HTTP requests
  • aiohttp.ClientSession for connection reuse
  • asyncio.gather() to fetch multiple URLs at once
  • Comparing sync vs async request times

Why this project exists

The real power of async shows up when you need to fetch data from many sources. Instead of waiting for each response before making the next request, you fire them all at once. This project makes that difference concrete.

Run

cd projects/modules/05-async-python/02-concurrent-requests
pip install -r ../requirements.txt
python project.py

Expected output

--- Sync requests (one at a time) ---
Fetched post 1: "sunt aut facere..." (270 chars) in 0.3s
Fetched post 2: "qui est esse..." (230 chars) in 0.2s
...
Sync total: ~2.5 seconds for 10 requests

--- Async requests (all at once) ---
Fetched post 1: "sunt aut facere..." (270 chars) in 0.3s
Fetched post 3: "ea molestias..." (250 chars) in 0.3s
...
Async total: ~0.4 seconds for 10 requests

Alter it

  1. Increase from 10 to 50 URLs. How do the sync vs async times compare now?
  2. Add a semaphore (asyncio.Semaphore(5)) to limit concurrent requests to 5 at a time.
  3. Fetch from two different API endpoints in the same gather() call.

Break it

  1. Remove async with session: and just use session = aiohttp.ClientSession(). What happens?
  2. Fetch a URL that does not exist. How does error handling differ from sync?
  3. Set a very short timeout (0.001 seconds). What exception do you get?

Fix it

  1. Add try/except around each fetch so one failure does not crash the whole batch.
  2. Add asyncio.wait_for() with a timeout per request.
  3. Close the session properly in a finally block if you removed the async with.

Explain it

  1. Why is aiohttp needed instead of just using requests with asyncio?
  2. What does ClientSession do that individual requests do not?
  3. How does a semaphore help when making hundreds of requests?
  4. What happens if you await each request individually instead of using gather()?

Mastery check

You can move on when you can: - use aiohttp to fetch multiple URLs concurrently, - explain why requests (sync) cannot be used with asyncio effectively, - add error handling per-request in a batch, - use a semaphore to limit concurrency.

Next

Project 03 — Async File Processing