Library
Content Catalog
Modules organized by source. Open any ready module to play it; coming-soon items are listed for reference. Press ⌘K anywhere to fast-switch.
System Design Interview Vol 1
BookAlex Xu · 2020Chapter 1
readyScale to Millions
Build an architecture that scales from one server to millions of users, stage by stage.
~20 min
Chapter 2
readyBack-of-the-Envelope Estimation
Reference cards and interactive calculators for the numbers every architect should be able to estimate quickly: data volumes, operation latencies, availability tiers, and a worked QPS-and-storage example.
~15 min
Chapter 3
readyInterview Framework
The 4-step framework Alex Xu prescribes for system design interviews, walked through end-to-end on the news-feed reference design — and a live Mock Interview Mode that role-plays the framework with a soft timer, contextual hints, and a wrap-up comparison against the book's reference design.
~60 min
Chapter 4
readyRate Limiter
Five algorithms for capping traffic — token bucket, leaking bucket, fixed window, sliding window log, sliding window counter — with their pros, cons, memory profiles, and where each fits in production. Plus the Redis-backed architecture, distributed-correctness concerns (race conditions, multi-instance synchronization), and HTTP 429 semantics.
~25 min
Chapter 5
readyConsistent Hashing
Why naïve hash-mod-N redistributes most keys when the cluster changes, and how the hash ring fixes it. Includes the rehashing-problem demo, the hash ring with clockwise lookup, add/remove animations with affected-arc highlighting, the two distribution issues, and virtual-node smoothing — plus a live backend stress test in the upcoming Live Lab section.
~30 min
Chapter 6
readyDesign a Key-Value Store
From a single-machine hash table to a Dynamo/Cassandra-shaped distributed store. Walks the CAP tradeoff, data partition & replication, the N·W·R quorum math, vector-clock conflict resolution, gossip / hinted handoff / Merkle anti-entropy, the peer-ring architecture, and the LSM-tree storage engine — plus two backend-backed Live Labs (Quorum Playground, Merkle Tree Sync).
~45 min