Command Palette

Search for a command to run...

Hectal

Database design · from first entity to FAANG-level architecture

Don't start with "which database?".
Start with the data, the access patterns, and the failures.

Database design taught the way senior engineers practise it: model entities and keys, normalize and denormalize deliberately, write serious SQL, reason about transactions, isolation, locks and MVCC, understand PostgreSQL internals (pages, WAL, VACUUM, B-trees vs LSM), design indexes from access patterns and read EXPLAIN ANALYZE, then scale with replicas, partitions and shards, pick NoSQL stores for the right reasons, keep copies in sync with outbox and CDC, and run it all in production: migrations, backups, DR, security, multi-tenancy, observability, capacity and cost. Every topic works through an interview problem and the ways it breaks, ending with eight domain designs and a full interview bank.

Phases
18
Topics
88
Hours
12
Diagrams
22
Exercises
299

By the end you can

What this course makes you able to do

  • Turn business requirements into entities, keys, constraints and a normalized schema, and denormalize on purpose
  • Write complex SQL (joins, window functions, recursive CTEs) and diagnose slow queries with EXPLAIN ANALYZE
  • Choose isolation levels and locking to prevent lost updates, write skew and deadlocks
  • Explain MVCC, WAL, VACUUM, B-trees and LSM trees, and tune PostgreSQL from its metrics
  • Scale with replication, partitioning and sharding, and pick Redis, MongoDB, Cassandra, DynamoDB or Elasticsearch for the right access patterns
  • Run databases in production: zero-downtime migrations, backups and PITR, DR, security, multi-tenancy, capacity and cost