Database

This section is about what happens inside a relational database: how rows are stored and found, how concurrent changes stay consistent, how the optimizer turns SQL into a plan, and how all of that changes once one machine is no longer enough.

The material is organized by the questions you ask in practice:

• How should I model the data? — design
• How does the engine actually find rows? — indexing
• How do concurrent changes stay consistent? — transactions
• Why is this query slow, and how do I fix it? — optimization
• What changes when one machine is no longer enough? — advanced

Topics

Modeling

Deciding what tables exist and how they relate before the engine ever sees a query.

• Design — Schema design, normalization assessment, data integrity, performance metrics, best-practices checklist.

Engine Internals

What the database is doing on every read and every write.

• Indexing — Index principles: B-Tree, hash, composite indexes, and the strategies that decide which one to use.
• Transactions — ACID properties, isolation levels, concurrency control, locking.

Performance & Distribution

Making slow queries fast, then keeping things working once one node is not enough.

• Optimization — Query optimization, execution-plan analysis, caching strategies.
• Advanced — Distributed databases, sharding, replication, CAP theorem.

Scope and Boundaries

The split with infrastructure/ is the most common point of confusion: this section is about what happens inside a relational database engine. Anything else that stores bytes — caches, blobs, search, vectors, warehouses, streams — lives in infrastructure/, because operationally it behaves more like infrastructure than like SQL.