Keep Building Fresh: Shopify's Journey to Kubernetes
summary
Shopify moves a monolith and hundreds of apps onto Kubernetes over about a year, proving a platform approach that hits ~80% of use cases and rolls changes through tiers to limit blast radius. Engineering builds servicesTP/croncontrol/people/deploy to create namespaces, scaffold runtimes, build images and run safe kubectl workflows, and writes 20+ custom controllers that auto-create DNS, TLS, databases and quotas. The platform handles ~80k req/s for core services, automates cluster creation and certificate/DNS plumbing, and uses staged rollouts and tests to catch controller regressions.
AI-written summary · about
chapters (9)
- 0:00Context and company scale
- 1:00Challenges of legacy systems
- 4:29Defining platform principles
- 8:16Choosing a solid foundation
- 14:56Building application blocks
- 19:54Deploying with automation
- 21:52Custom Kubernetes controllers
- 34:38Developer support and resources
- 37:27Lessons learned and future
related talks
33mBuilding Shopify's PaaS on Kubernetes
19mBuilding an On-Premise Kubernetes...
41mShopify's Move from the Data Centre to the Cloud
04mKeynote: One Deploy Plane, Every Runtime: How Shopify Generalized Cloud Native Beyon...
32mOperating Rails in Kubernetes
40mSecuring Shopify's PaaS on GKE [I]
44mMigrating a Monolith to Kubernetes
32mScaling Shopify's Search: Enhancing Elasticsearch Resilience With Kubernetes and KE...