Choreography vs Orchestration in serverless microservices
summary
Compares event-driven choreography with a central orchestrator by reimplementing a photo‑sharing demo: the original Pub/Sub + Cloud Functions + Cloud Storage + Vision API + Firestore flow was replaced by a Google Cloud Workflows YAML that calls Vision API, Firestore, Cloud Storage and Twilio as declarative steps. Finds that orchestration simplifies monitoring and correlating runs, centralizes retries and timeouts, and reduces glue code and some services, while introducing a single orchestration dependency and making dynamic event subscription harder.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:01Orchestration vs choreography defined
- 2:49Synchronous approaches limits
- 5:22Async event choreography
- 7:11Choreography pros and cons
- 11:08Central orchestrator
- 16:46App demo
- 17:50Orchestrator internals
- 23:54Orchestration vs choreography
- 26:18Available solutions
- 28:27Google Cloud Workflows deep dive
- 31:18Advanced orchestrator features
- 37:15Interface demo
- 37:57Photo-sharing app case study
- 41:08Rewrite experience
- 44:40Limits and best practices
- 45:45Resources and conclusion
related talks
01h 03mChoreography vs Orchestration in serverless microservices
51mChoreography vs Orchestration in serverless microservices
50mChoreography vs Orchestration in serverless microservices
49mChoreography vs Orchestration in serverless microservices
46mChoreography vs Orchestration in serverless microservices
32mChoreography vs Orchestration in serverless microservices
38mChoreography vs Orchestration in serverless microservices
50mChoreography vs Orchestration in serverless microservices
