Performing Infrastructure Migrations at Scale
summary
Airbnb migrates hundreds of legacy services to Kubernetes, reaching 70% of services in Kubernetes and running 300 critical services in production after a 1–2 year effort. The program builds infrastructure-as-code, a generator to make new services default, and refactor bots that create and merge PRs to automate upgrades (security image bumps, CI conversions). A dedicated migration team sequences work, stress-tests hard cases, and reports 70% automated completion while the remaining 30% require manual, one-off fixes.
AI-written summary · about
chapters (21)
- 0:00Intro
- 0:53Migrations: Airbnb Case Study
- 1:25Example Migrations
- 4:46Migrations reduce tech debt
- 5:21Examples of tech debt
- 6:31Migration types
- 9:27Cascading Migrations
- 11:02Simultaneous Migrations
- 13:27Sequenced Migrations
- 16:02Prioritized Migrations
- 17:55Migration at scale
- 24:39Migration overhead: what developers get
- 25:10Unfinished migrations
- 27:36better abstraction?
- 29:26How do we migrate programmatically?
- 29:49The lifecycle of a refactor
- 31:06What do we migrate programmatically?
- 31:54Migration strategy: make one person do it
- 32:15Migration strategy: make devs do all of it
- 32:37Migration strategy: an actual migration program
- 34:2610 Takeaways
related talks
35mPerforming Infrastructure Migrations at Scale
14mAI-accelerated migrations at scale
48mScaling a Migration to CD at Airbnb
38mBuilding the Perfect Infrastructure with Kubernetes
08mOpening Keynote
39mLessons Learned From Maintaining Continuous Delivery While Migrating From a Static Infrastructure to Kubernetes [B]
44mMigrating a Monolith to Kubernetes
34mKubernetes from Dev to Prod at GoEuro