Kubernetes-as-a-Service lets a cloud provider manage the Kubernetes cluster and its components while users deploy applications through the Kubernetes API. It separates using Kubernetes from the difficult work of operating a production-ready cluster. section 1.3.2 section 1.4
Choose a managed offering when you want to:
- Avoid taking responsibility for the cluster’s operational complexity and maintenance. section 1.3.2
- Deploy applications with the Kubernetes API and kubectl rather than building the cluster yourself. section 1.3.2 section 3.4
- Use familiar Kubernetes resources such as Deployments, Services, Pods, and Nodes. section 3.4
Typical managed-cluster workflow
After obtaining a managed cluster, use kubectl to create a Deployment, expose it with a Service, and scale it by changing the desired replica count. A LoadBalancer Service can request a publicly accessible address from the surrounding cloud infrastructure. Common offerings include Google Kubernetes Engine, Azure Kubernetes Service, and Amazon Elastic Kubernetes Service. section 3.4 section 1.3.2
