Things I Wish I knew Before I Started with Event Sourcing
summary
Stores every state change as an append‑only event log and shows how that becomes the single source of truth, with concrete choices: use Postgres or a dedicated event store, prefer ProtoBufs for evolvable payloads, avoid treating Kafka as a per‑aggregate journal, and build a small CLI snapshots + replay debugger. Covers snapshotting and single‑writer/cluster sharding for hot aggregates, compensating events instead of manual DB edits, CDC or polling read‑sides, and when to accept eventual consistency versus forcing strong consistency.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:05Event sourcing basics
- 8:55Need a framework?
- 15:03Time travel debugging
- 18:59Never edit streams
- 24:02Fixing projections
- 25:32Powering insights and future business questions
- 28:46Eventual consistency
- 33:42Strong consistency cases
- 34:04Performance and snapshots
- 36:34Domain modeling
- 38:24Single writer principle
- 41:57System maintenance
- 44:04Event versioning
- 44:26Read side scaling
- 45:31Private vs public events
- 46:12Summary
related talks
48mEvent Sourcing - what could possibly go wrong?
52mEvent Sourcing for the Cloud Developer
39mEvent Sourcing with Elixir
43mUnraveling the Threads: Event-Sourcing, CQRS, and Event-Driven Architecture
52mEvent Sourcing – the what, why & how
38mEvery Event, Everywhere, All at Once
10mEvent sourcing
51mMicroservices Data Patterns: CQRS & Event Sourcing