System Design Fundamentals

System Design Fundamentals

System design is the process of defining the architecture, components, interfaces, and
data flow of a software system to satisfy a set of functional and non-functional
requirements — things like scale, latency, availability, and cost.

Why it matters

• Real engineering. Every production system that serves more than a handful of users
  eventually has to answer questions about scale, failure, and change.
• Interviews. Most senior/staff engineering interviews at large tech companies include
  a system design round, because it's a good proxy for how you think, not just what you know.

A repeatable approach

1. Clarify requirements. Separate functional requirements (what the system does) from
   non-functional ones (scale, latency, consistency, availability, cost).
2. Estimate scale. Back-of-the-envelope math for traffic, storage, and bandwidth — see
   Back of the Envelope Calculation.
3. Sketch a high-level design. Boxes and arrows: clients, load balancer, services,
   datastores, caches, queues.
4. Go deep on 1–2 components. Whatever is most interesting or most constrained —
   usually the data model or a specific bottleneck.
5. Discuss trade-offs. There is rarely one right answer; the value is in reasoning
   explicitly about the trade-offs you're making.

Where to go next

Read System Design Principles for the recurring rules
of thumb, then System Design Concepts for the vocabulary
you'll need for everything else in this handbook.