Going Multi Cloud in a Hurry with Quality and Style
summary
A GKE-based managed Kubernetes platform adds AWS by standing up an EKS cluster, reusing Argo CD and Helm, and implementing Prometheus, Grafana, L7 load balancers and secrets management; a five-person cross‑functional team with no AWS experts delivers a production-quality platform in 6 months despite an initial five-week ask. Higher compliance forces upfront diagrams and documentation, so iterations include dedicated compliance sign‑offs, training via recurring jam sessions, and tool-driven deployments with Argo CD; system runs a year with no major incidents.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:25Managed Kubernetes in GCP
- 3:10Team setup and AWS gap
- 4:32Compliance requirements and timeline
- 5:56Adjusting plans and communication
- 7:00Jam sessions for learning
- 9:08Jam session structure
- 10:11Refining jam sessions
- 11:15AWS platform ahead of schedule
- 12:41Psychological safety
- 13:43Soft skills matter
related talks
33mKubernetes-Driven Multi-Cloud Deployments: Maximizing Scalability and Flexibility
35mKubernetes journey
36mMulti-Cluster Kubernetes Quick Delivery with IaC and GitOps
27mKubernetes Everywhere: Lessons Learned From Going Multi-Cloud
32mGitOps, an Elegant Tool for Hybrid Cloud Kubernetes
06mArgo CD and Kubernetes from event based to microservices
35mMulti-Cluster Management Made Easy - Tsahi Duek and Dima Breydo
36mThe Architecture of a Multi-Cloud Environment with Kubernetes