Everything Loosely Coupled - Advanced Integration Patterns and Their Trade-Offs
summary
Practical claim: switching synchronous APIs to message-based async with queues reduces temporal coupling and avoids blocked resources; examples use Amazon SQS (virtually infinite scale) and Apache ActiveMQ (instance limits). Concrete patterns shown: message group IDs to preserve order within groups at cost of concurrency, dead-letter queue for poison pills, correlation IDs and return-address for replies, and archive+replay via SNS/EventBridge requiring idempotent consumers. Measured trade-offs: FIFO ordering lowers throughput, retries can cascade failures, and exactly‑once semantics force shared state or synchronous designs.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 2:45Monolithic Systems
- 6:17Integration isnt an afterthought
- 8:35Integration takes many shapes
- 10:18No architecture decision comes without tradeoffs
- 13:38Message Cues
- 16:49Example
- 21:25Downstream Services
- 25:34Queues
- 30:40Order
- 37:19Duplication
- 40:15Poison Pill
- 42:06Archive Replay
related talks
46mIntegration patterns & trade-offs for loosely coupled systems
48mIntegration- and conversation-patterns for microservices
44mThinking Asynchronously: App Integration Patterns for Microservices
01h 01mBeyond Pub/Sub - Advanced Messaging Patterns
01h 01mThe Unsung Hero of Modern Software: Asynchronous Messaging
53mYou Keep Using That Word: Asynchronous And Interprocess Comms
48mI Made Everything Loosely Coupled. Does My App Fall Apart?
42mIntegration and Conversation Patterns in SaaS Applications