CQRS pitfalls and patterns

on YouTube ↗59m

" summary

Claim: applying CQRS blindly often increases complexity and creates user-visible bugs rather than improving performance. The talk contrasts standard single-database CQRS with eventually consistent CQRS (commands write to one DB, events update a separate query DB), shows how an async query updater and optimistic concurrency produce stale or duplicated records under load (200 concurrent buys against 1,000 inventory became 999), and recommends alternatives: DB replication/sharding, explicit private vs public data flows, pre-charge or negative-inventory business models, and vertical-slice composition.

AI-written summary · about

" chapters (16)

1. 0:00Introduction
2. 1:00CQRS Pitfalls
3. 4:41Best Practices
4. 7:10Command Query Separation
5. 10:23Domain Models
6. 13:21Architecture
7. 16:37Standard CQRS
8. 20:39Simple CQRS
9. 25:18Private vs Public Data
10. 31:22Problems with CQRS
11. 34:14Vertical Slice Architecture
12. 37:49Race Conditions
13. 43:17Optimistic concurrency
14. 45:46Connection pool
15. 52:42Nonblocking
16. 55:38Advanced composition

" related talks

• 31mCQRS: Do you really need two databases?Dejan Miličić · Øredev 2023
• 04mCommand Query Responsibility SegregationOlivier Loverde · dotJS 2018
• 52mGoing eventually consistent with CQRSManel Sellés · Software Craftsmanship Barcelona 2017
• 17m7 Reasons why your microservices should use Event Sourcing & CQRSHugh McKee · Voxxed Days Microservices 2019
• 46mMicroservices Data Patterns: CQRS & Event SourcingEdson Yanaga · Voxxed Days Bucharest 2018
• 50mMoving away from CRUD: DDD + CQRSForat Latif · Codemotion Roma 2014
• 40m7 Reasons why your microservices should use Event Sourcing & CQRSHugh McKee · Voxxed Days Cluj-Napoca 2019
• 55m7 Reasons why your microservices should use Event Sourcing & CQRSHugh McKee · Riga Dev Days 2019