Every system wants to be fast, scalable, and reliable, which is exactly why saying so is worthless. This post is about the sharper version: p99 under 200 ms, 99.99% availability, eventual consistency on the timeline — and the architectural lever each of those numbers forces you to pull.

Most system design answers go wrong one step before the boxes and arrows: the candidate never works out how big the system is. Here are the four formulas, the latency numbers worth memorizing, and a worked Twitter example that turns four calculations into five architecture decisions.