Go micro with Microservices
summary
Shows extracting parts of a monolith (recommendation engine scaled to 1 million pageviews in a week) into independent services built with PHP 7 and Silex/Symfony while using RabbitMQ/AMQP for async communication, composable UIs or an API gateway for aggregation, and caching with Varnish. Covers operational needs: service discovery (ZooKeeper), correlation IDs for distributed logs, circuit breakers, and monitoring with InfluxDB + Grafana. Warns that preparation can take months — one company spent six months and shipped one trivial microservice — and that eventual consistency is inevitable.
AI-written summary · about
chapters (21)
- 0:00Start
- 1:42THE STORY
- 6:11REASONS...
- 8:00LOOSELY COUPLED SERVICE ORIENTED ARCHITECTURE WITH BOUNDED CONTEXTS
- 12:30INTEGRATION
- 16:20UI COMPOSITION ESI/SSI/AJAX/IFRAME
- 16:47COMPOXURE MIDDLEWARE FOR EXPRESS
- 18:42API GATEWAY
- 19:49BACKEND FOR FRONTEND
- 20:09STRUCTURING
- 24:26YOU NEED A SKELETON
- 26:22MILTON FRIEDMAN THERE AIN'T NO SUCH THING AS A FREE LUNCH
- 26:49DISTRIBUTED COMPLEXITY
- 27:11COMMUNICATION
- 28:22HTTP VS NO-HTTP
- 34:03MONITORING
- 36:07SERVICE DISCOVERY
- 36:45DATA STORES
- 37:45EVENTUAL CONSISTENCY
- 39:09ONE SERVICE MAY RUIN EVERYTHING
- 40:59ARE MICROSERVICES FOR ME...?
related talks
52mMicroservices: A Utopian Mystery
48mChallenges in Implementing MicroServices
34mLevel up with a Microservices architecture
57mUnderstanding Microservices: a guide for the monolithic developer
54mThe Long Sad History of Microservices
50mChallenges in Implementing Microservices
24mPatterns of Evolution from Monolith to Microservices
01h 01mMicroservices Gone Wrong
