The Paved PaaS to Microservices
summary
Netflix describes a concrete PaaS that fronts hundreds of microservices with the Netflix Edge API, enforces a single RPC and standardized components (metrics, logs, discovery, Falkor) to cut MTTR, and supports thousands of A/B tests per year by letting teams push code daily. The platform is delivered as preassembled layers, a CLI that bootstraps dev environments and local Docker work, semantic versioned releases with conventional changelogs, built-in mocks and test APIs, and automated canary deploys and dashboards.
AI-written summary · about
chapters (34)
- 0:00Intro
- 2:14Microservices
- 5:37Goals
- 6:19Know Your Customer
- 7:20What's Inside a Microservice?
- 7:56Why Standards?
- 8:32Microservice Interactions
- 8:57Mean Time to Detection (MTTD) Mean Time to Repair (MTTR)
- 9:39One Standard RPC
- 9:51Benefits of Standardizing
- 10:46But I'm a Snowflake...
- 12:11Assembly Required
- 12:43Getting out of the Blocks
- 14:44Application, system, and runtime metrics & logs
- 15:13Consistent application system, and runtime metrics & logs
- 15:38Configures and Initializes Correctly
- 17:25Maintenance vs Convenience
- 17:59Solution: Layers & Flavors
- 19:38Component Correctness
- 20:44Tradeoffs
- 21:26Season to Taste
- 22:16API Semantic Versioning
- 23:04Ship a Feature
- 23:17Steps
- 24:18CLI for common dev experience
- 26:06Local Development
- 27:50Provide First Class Mocks
- 28:14Platform Testing API
- 29:28Experience Differences
- 29:55Deploy and Manage Services
- 31:33Consolidated View
- 31:53Generated Dashboards & Alerts
- 32:19Automated Analytics & Tooling
- 32:39Our PaaS to Velocity & Reliability
related talks
50mFast Delivery
20mThe Business Benefits of Microservices
08mThe Evolution of Microservices Architecture in Cloud-Native
25mMaking microservices a behaviour not a buzzword
26mMicroservices
08mMicroservices at Scale
42mInnovating Faster with Micro-Service Architecture...
50mMicroservices Design Patterns for Java Application