Blog

What an index physically is, how B-trees make lookups logarithmic, composite index ordering, covering and partial indexes, reading EXPLAIN ANALYZE, index types beyond the B-tree, adding indexes safely in production, and the query patterns that silently refuse to use your index.

How load balancers decide where each request goes — the classic algorithms, L4 vs L7, health checks, and the sticky-session problem that pushes state out of your servers.

Why `hash % N` falls apart the moment you add a server, how consistent hashing moves only a fraction of your keys when the cluster changes, virtual nodes, replicas on the ring, bounded loads, rendezvous and jump hashing, and what Cassandra and Redis Cluster actually do.

Why the classic interview question is really a lesson in composing data structures — with an animated walkthrough and two working implementations.

How APIs decide who gets throttled — token bucket, fixed and sliding windows, animated demos of each, and what changes when the limiter goes distributed.

Four ways to get server data to the browser without a refresh — animated message flows for each, and a decision guide drawn from building live-data UIs.

Build a micro-frontend architecture with Next.js multi-zones: rewrites, assetPrefix, cross-zone navigation, shared auth, deployment, and the pitfalls nobody mentions.