Kubernetes is an open-source system for automating the deployment, scaling, and management of containerized applications. It groups containers into logical units for easy management and discovery, and continuously works to keep the actual state of a cluster matching the state you declare.

How it works

A Kubernetes cluster is made up of a control plane (API server, scheduler, controller manager, etcd) and worker nodes running a container runtime and the kubelet agent. You describe the desired state — how many replicas of a Pod should run, what resources they need, how they should be exposed — as objects in the API, and Kubernetes’ controllers continuously reconcile the cluster toward that state.

Core concepts

• Pod — the smallest deployable unit: one or more containers that share network and storage.
• Deployment — manages a set of replica Pods and handles rolling updates and rollbacks.
• Service — a stable network endpoint that load-balances traffic across a set of Pods.
• Namespace — a way to partition a cluster into isolated logical environments.
• Controller — a control loop that watches the cluster’s state and moves it toward the desired state.

Why teams adopt Kubernetes

• Portable. The same manifests run on any conformant cluster, from a laptop to a large multi-region deployment.
• Self-healing. Failed containers are restarted, unhealthy nodes are drained, and workloads are rescheduled automatically.
• Extensible. Custom Resource Definitions and controllers let you extend the API to manage your own domain-specific resources.
• Ecosystem. Nearly every CNCF project — service meshes, observability tools, GitOps controllers — is built to run on and extend Kubernetes.

Getting started

Install kubectl, get access to a cluster, and apply a Deployment and Service manifest to run your first workload. The guides below build on that foundation with common patterns for running real applications.