Fundamentals

How systems grow with load — vertical vs horizontal scaling — and how load balancers spread traffic across servers.

Storing frequently used data in fast storage to cut latency and load — cache strategies, eviction, and invalidation.

Choosing between relational (SQL) and non-relational (NoSQL) databases based on schema, consistency, and scale needs.

Splitting data across nodes (sharding) for scale and copying it (replication) for availability and read throughput.

In a network partition a distributed store must choose between consistency and availability — the CAP trade-off.

A hashing scheme that minimizes key remapping when nodes are added or removed — key to distributed caches and shards.