Beyond the current state: Time travel as answer to hard questions
summary
Explains how splitting writes and reads with CQRS plus storing every change as immutable events with event sourcing lets a system replay history to reconstruct state, implement time travel for exact debugging, and rebuild projections in any storage (relational, in‑memory, GraphQL layers). Covers practical details: return an ID immediately, make projectors idempotent, use aggregate sequence numbers for concurrency, map commands one‑to‑one to events, and apply Saga patterns for non‑idempotent third‑party interactions; notes trade‑offs in hiring and eventual consistency.
AI-written summary · about
chapters (8)
- 0:00Introduction and speaker intro
- 1:05Traditional three-tier architecture
- 2:30Limitations of current state approach
- 5:26CQRS and event sourcing basics
- 10:09Event sourcing and time travel
- 13:03Superpowers of event sourcing
- 18:33Challenges and trade-offs
- 22:54Conclusion and Q&A
related talks
56mDeveloping Event-driven Microservices w/Event Sourcing and CQRS
56mAn Abridged Guide to Event Sourcing
47mGive Your Application Memories
43mFunctional and Reactive Event Sourced Applications
52mGoing eventually consistent with CQRS
52mArchitectural Patterns for Distributed and Decentralized Data
40m7 Reasons why your microservices should use Event Sourcing & CQRS
46mState or events? Which shall I keep?