Fast Delivery
summary
Describes moving Netflix to AWS, building a microservice platform with Cassandra, Docker/AMIs, and a 200 terabyte SSD cluster used for quick infrastructure experiments; teams scale to ~5–600 engineers and hundreds of services (Gilt hits ~450), deploy immutable releases, use NetflixOSS tools like Hystrix/Turbine for circuit breaking and per-second dashboards, and run continuous delivery with feature toggles and A/B tests to cut decision-to-deploy time by orders of magnitude.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:22Typical reactions to my Netflix talks...
- 3:53What I learned from my time at Netflix
- 9:10Cloud Adoption
- 16:19Process Hand-Off Steps for Product Development on laas
- 17:30Process Hand-Off Steps for Feature Development on Paas
- 18:47Saas Based Business Application Development
- 24:11Non-Destructive Production Updates
- 25:46What Happened?
- 28:48Separate Concerns with Microservices
- 31:35NetflixOSS - High Availability Patterns
- 32:22Open Source Ecosystems
- 33:06Microservices Development
- 35:45Microservice Datastores
- 38:37Cloud Native
- 39:49Microservice Based Architectures
- 40:12"Death Star" Architecture Diagrams
- 41:19Continuous Delivery and DevOps
- 43:02Netflixoss Hystrix/Turbine Circuit Breaker
- 44:07Low Latency SaaS Based Monitors
- 45:29Forward Thinking
- 48:12Any Questions?
related talks
29mHow Netflix Thinks About DevOps
31mHow Netflix Thinks About DevOps
29mFast, Safe, and Continuous -- Picking all 3 at Netflix
31mKeeping Your Cloud Footprint in Check
58mBuilding a Culture of Continuous Delivery
46mCassandra Serving Netflix @ Scale
27mBehind the Scenes: Running Critical Services @ Netflix
08mMicroservices at Scale