Async/Await Demo: API + Cache + DB

This project is a teaching demo that shows what async/await buys you in Python.

It starts 3 fake servers in one process: - API server (127.0.0.1:9000) - Cache server (127.0.0.1:9001) - Database server (127.0.0.1:9002)

The API receives a client request and needs several pieces of data: - user profile (cache, then DB fallback) - product info (cache, then DB fallback) - inventory (DB) - recommendations (DB)

It does this in two ways: 1. sequential: await each call one by one 2. concurrent: start independent calls together with asyncio.create_task(...) and await asyncio.gather(...)

The script runs both modes and prints timing so the speed difference is obvious.

Run

uv run demo_async_await.py

What to watch for

• Timestamped log lines from API, CACHE, and DB
• In sequential mode, DB calls happen one after another
• In concurrent mode, calls overlap and complete sooner
• Same final data, lower total latency

Last Updated 02/17/2026