Kubernetes
Kubernetes (often called K8s) is an open-source platform for automating deployment, scaling, and management of containerized applications.
It provides a powerful, declarative API so you can describe the desired state of your workloads and let the system maintain it.
Overview
- Container Orchestration – schedules and runs containers across a cluster of machines.
- Declarative Configuration – you describe the desired state (e.g., “3 replicas”) and Kubernetes continuously works to match it.
- Self-Healing – automatically restarts or reschedules failed containers.
- Horizontal Scaling – add or remove pods seamlessly based on demand.
Architecture
The diagram below shows the main components of a Kubernetes cluster:
Key Pieces
Control Plane
The control plane manages the entire cluster and keeps the system’s desired state.
| Component | Role |
|---|---|
| API Server | Central management entrypoint; validates and processes requests. |
| Scheduler | Chooses which node should run a new pod. |
| Controller Manager | Ensures the cluster matches the declared state (replicas, endpoints, etc.). |
| etcd | Highly available key–value store holding cluster configuration and state. |
Worker Nodes
Worker nodes run your actual workloads (containers/pods).
| Component | Role |
|---|---|
| Kubelet | Agent on each node that talks to the API server and manages pods. |
| Kube-Proxy | Maintains networking rules and handles service discovery/load-balancing. |
| Container Runtime | Software (e.g., containerd, CRI-O) that runs the containers. |