Managing OS Release Transitions at Netflix Scale
summary
Netflix migrates over 150k+ Ubuntu instances across multiple AWS accounts by building a controlled BaseAMI pipeline and a Bakery that produces immutable application AMIs, automated nightly builds and monthly releases, and daily microbenchmarking across every instance type so OS regressions are caught early. Developers adopt a meta-config tool easy config to avoid init changes, canary groups and a mix of nightly candidate and monthly release branches enable rapid Xenial→Bionic turnover.
AI-written summary · about
chapters (27)
- 0:00Intro
- 0:56painless transition
- 1:13Netflix
- 2:04Amazon Cloud
- 2:32Hardware Configurations
- 3:14Microservices
- 4:12Innovation
- 4:50Dont Crash
- 5:25Route Planning
- 6:01Constraints
- 6:38Webstore
- 7:00Prealpha users
- 7:27Base AMI
- 8:49Developers
- 9:15EasyConfig
- 10:36Putting it all together
- 10:49Standard release process
- 11:18Unstable base
- 13:08Automation
- 14:22Tracking Progress
- 15:05Communication
- 16:35Unicorns
- 16:46Account Structure
- 17:43Organizational Structure
- 19:07Timeline
- 20:11People
- 22:09Questions
related talks
25mJourney to Automated Releases
29mHow Netflix Thinks About DevOps
31mHow Netflix Thinks About DevOps
50mFast Delivery
45mNetflix’s Secret to Safe Automation at Scale
27mBehind the Scenes: Running Critical Services @ Netflix
58mBuilding a Culture of Continuous Delivery
31mKeeping Your Cloud Footprint in Check