Systems & Scale

What breaks when one machine becomes many, and when two things happen at once.

12 concepts · 495 minutes

• CDNs & the EdgeMove the bytes closer to the user — the highest-leverage latency fix available.
• The Node.js RuntimeJavaScript outside the browser: modules, streams, and a single thread doing a lot of I/O.
• IdempotencyMaking 'do it again' safe — the single idea that separates reliable distributed code from hope.
• HTTP CachingCache-Control, ETags, and revalidation — the cheapest performance win that exists.
• Caching with RedisAn in-memory store used as cache, session bag, lock, counter, and queue.
• Load BalancingSpreading traffic across machines, and what breaks the moment there's more than one.
• Scaling Out & StatelessnessWhy your server must not remember anything locally if you want a second one.
• The Event LoopOne thread, a queue, and a rule about when it drains — the model that explains every 'why did this run last?'.
• Concurrency vs ParallelismDealing with many things at once vs doing many things at once — not the same, and the distinction matters.
• CAP & ConsistencyDuring a partition you pick consistency or availability — everything else is marketing.
• System Design InterviewsClarify, estimate, sketch, then defend the trade-offs — a repeatable structure for an open-ended hour.
• Race ConditionsTwo things interleaving in an order you never tested — the bug class that only shows up in production.