DEVOPS DAYS MTL 2018 - Maxime Boisvert
summary
Moves a self-hosted Elasticsearch cluster to Google Cloud using k8s and Go and reports concrete outcomes: defines SLO from end-user perspective, maps each Elasticsearch node = k8s pod, builds a custom controller in Go to automate provisioning, and executes a parallel migration with rollback plan. Sees cloud flakiness and I/O throttling require stronger resilience, dealings with cloud support, and lost low-level control. Gains faster onboarding, infrastructure-as-code, reduced toil, and more focus on core search reliability.
AI-written summary · about
related talks
33mKubernetes-Driven Multi-Cloud Deployments: Maximizing Scalability and Flexibility
35mThe Oregon Trail to Kubernetes [I]
35mKubernetes journey
24mFrom Lift-and-Shift to K8s: A Two-Phase DevOps Modernization
31mAn Adventure in Multi Cloud with Kubernetes and Distributed Data
02h 18mDevelop and Deploy to Kubernetes like a Googler
24mA Dash of Bazel
37mBalancing Power and Pain: Moving a Startup From a PaaS to Kubernetes