This talk was given at KubeHuddle in Edinburgh.
Monzo had been running a single self-hosted Kubernetes cluster per environment for years, and we decided to migrate to a managed Kubernetes service. This talk covers the challenges and perils of doing a high-stakes, real-world, zero-downtime Kubernetes migration at scale.
We talked about the constraints we operated under: huge diversity in the systems running on the cluster, downtime having real human impact, and the number of teams involved, which meant we needed an automated process that was fully transparent to other engineers.
The takeaways covered how Kubernetes can make infrastructure migrations more manageable and actionable learnings about doing migrations at scale in general.
Session details are available on Sessionize.
