Kubernetes Control Plane Diagram

Kubernetes Control Plane and Worker Nodes

This Kubernetes architecture diagram separates the control plane from the worker nodes. The API server is the central control-plane interface. It communicates with etcd for cluster state, with the scheduler for placement decisions, and with the controller manager for reconciliation. The API server also communicates with two worker nodes through their kubelets. Each worker node runs application pods. The diagram is intentionally high-level, making it suitable for onboarding or documentation. Add networking, ingress, storage classes, or an external control-plane endpoint when those parts matter to the design. Keep the diagram current when equipment, cable paths, service ownership, or security boundaries change, so it remains useful for planning and operations.

類型 網路圖標準 Cisco-convention topology icons引擎 schematex-network更新時間 2026/9/24

需求

“A Kubernetes cluster with an API server connected to etcd, the scheduler, controller manager, and two worker nodes; each worker node runs application pods.”

接著試試Add a kube-proxy on each worker nodeAdd a third worker node

這張圖裡有什麼

看懂背後的決策。

01Cluster state

etcd stores the control-plane state.

02Scheduling

The scheduler assigns pods to workers.

03Workloads

Worker nodes run the application pods.

Use this to introduce the Kubernetes control plane and the relationship between workers and pods.

瀏覽所有網路圖範本 →

常見問題

What is the Kubernetes control plane?

It is the set of components that manage the desired state of a Kubernetes cluster.

What does etcd store?

etcd stores Kubernetes cluster state and configuration.

What runs application pods?

Worker nodes run the pods scheduled onto them.