What Is Kubernetes? Container Orchestration Explained
Kubernetes is a container orchestrator: you declare the desired state of your apps, and it continuously works to make reality match.
Running one container is easy; running hundreds across many machines, keeping them healthy, scaling them, and rolling out updates without downtime is not. Kubernetes (often "K8s") automates that. You describe what you want in declarative manifests, and Kubernetes does the work of getting there and staying there.
Declarative, desired-state model
You do not tell Kubernetes how to act step by step. You declare a desired state ("run 3 replicas of this image"), and controllers continuously reconcile actual state toward it - restarting crashed containers, replacing failed nodes, and scaling as configured.
The control plane and nodes
The control plane (API server, scheduler, controllers, etcd) is the brain that stores state and makes decisions. Worker nodes run the kubelet and your actual container workloads. Together they form a cluster.
Core objects
- Pod - the smallest unit, one or more co-located containers.
- Deployment - manages replicas and rolling updates of pods.
- Service - a stable network endpoint for a set of pods.
- ConfigMap and Secret - configuration and sensitive data.
Why teams adopt it
Self-healing, horizontal scaling, rolling deploys, and a uniform API across clouds. The cost is operational complexity - Kubernetes has a real learning curve.
Kubernetes and CI/CD
Pipelines build images, then apply manifests (often via kubectl or GitOps) to deploy them to a cluster. CI also frequently builds the very images Kubernetes runs; managed runners keep those image builds fast and reliable.
Key takeaways
- Kubernetes is a declarative container orchestrator that reconciles to desired state.
- A cluster has a control plane (the brain) and worker nodes (the muscle).
- CI builds images; CD applies manifests to roll them out to the cluster.