Life Beyond Distributed Transactions: An Apostate's Implementation
summary
Implements Pat Helland’s “life beyond distributed transactions” with concrete code using Azure Cosmos DB (demo via emulator), RabbitMQ and the outbox pattern, plus examples noted for Azure Service Bus and Azure SQL Server. Builds per-document outbox/inbox in each document plus a dispatcher + retry queue and a unit-of-work that dispatches messages to target documents; uses inbox deduplication and a catch‑up trigger for missed dispatches. Demo shows a saga-based order fulfillment handling happy path, stock-confirm failures and rollbacks; it works but increases write volume and requires inbox cleanup.
AI-written summary · about
chapters (11)
- 0:00Introduction and background
- 2:29Distributed transactions problems
- 9:29Life beyond distributed transactions
- 11:55Inbox and outbox pattern
- 21:51Coordinating activities with messages
- 25:46Sagas for order fulfillment
- 29:25Live demo walkthrough
- 35:19Handling failure scenarios
- 46:15Other use cases and libraries
- 49:24Final lessons and summary
- 50:24Q&A
related talks
53mLife Beyond Distributed Transactions: An Apostate's Implementation
01h 02mLife Beyond Distributed Transactions: An Apostate's Implementation
01h 00mLife Beyond Distributed Transactions: An Apostate's Implementation
01h 02mConsistency and Agreements in Distributed Systems
53mEffective Microservice Communication and Conversation Patterns
01h 01mBeyond Pub/Sub - Advanced Messaging Patterns
01h 03mBeyond Pub/Sub - Advanced Messaging Patterns
01h 02mAt Least Once