Tools · SPBoucher
Menu

Kubernetes

In-depth entry

by Cloud Native Computing Foundation · Platform · also known as K8s

Open-source system for scheduling, scaling and operating containerized workloads across a cluster of machines

Apache-2.0 licenseOpen sourceFreeSelf-hostedCloudSelf-hostable
Visit official site Compare

Link verified Nov 2, 2026

Overview

Kubernetes is the container orchestration platform originally developed at Google and now a graduated project of the Cloud Native Computing Foundation, licensed under Apache-2.0. It turns a set of machines into a cluster on which declarative objects, such as Deployments, Services, ConfigMaps and PersistentVolumeClaims, describe the desired state; controllers continuously reconcile actual state toward it, restarting failed pods, rolling out new image versions and rescheduling work when nodes disappear.

It is used wherever applications span more than one host or need automated self-healing, horizontal scaling and service discovery. Managed offerings from every major cloud (EKS, GKE, AKS and others) remove control-plane operation, while distributions such as k3s or kind run on a single machine for development and edge use. The API is extensible through Custom Resource Definitions and operators, which is why databases, message brokers and machine-learning pipelines commonly ship Kubernetes-native deployment paths. Ingress controllers and the newer Gateway API bring traffic from reverse proxies like Traefik or NGINX into the cluster.

The platform's main cost is operational complexity. Networking, storage, secrets, RBAC and upgrades each require decisions and tooling, and the release cadence means clusters need regular maintenance. For a handful of services on one or two servers, Compose files or a process manager are often sufficient, and small teams frequently defer Kubernetes until workload count or availability requirements justify it.

Capabilities

  • Declarative deployments with rolling updates and rollbacks
  • Automatic rescheduling of failed pods and horizontal pod autoscaling
  • Service discovery, load balancing and Ingress or Gateway API routing
  • Persistent storage through volume plugins and the CSI interface
  • Role-based access control, namespaces and network policies
  • Extensibility through Custom Resource Definitions and operators
  • Helm charts and kustomize for packaging and configuration

Typical workflows

  1. Deploy a multi-service application with health checks and autoscaling
  2. Run batch or scheduled jobs across a node pool
  3. Operate a managed cluster on a public cloud
  4. Develop against a local single-node cluster

Strengths

  • Vendor-neutral API supported by every major cloud
  • Large ecosystem of operators, charts and tooling
  • Self-healing and scaling built into the control loop
  • Graduated CNCF project with a predictable release cycle

Limitations

  • Substantial operational complexity and learning curve
  • Regular upgrades are needed to stay within the supported version window
  • Networking and storage depend on cluster-specific plugins
  • Overkill for small single-host deployments

Prerequisites

  • Linux nodes with a container runtime (containerd, CRI-O) or a managed cloud offering
  • kubectl client and familiarity with YAML manifests

Who it is for

  • Platform and site-reliability engineers
  • Teams running microservices
  • ML engineers scheduling training or inference jobs

Alternatives & comparison

Docker Compose and Swarm cover single-host or small-cluster deployments with far less machinery, and Podman can generate Kubernetes YAML from local pods for a migration path. Platforms such as Fly.io provide scheduling and scaling as a managed service without exposing the Kubernetes API. Kubernetes is the choice when portability across providers and a rich operator ecosystem matter more than simplicity.

  • Docker

    Container platform combining the Docker Engine runtime, the Compose and Build tooling, Docker Desktop for macOS and Windows, and the Docker Hub registry

    MixedOpen source
  • Fly.io

    Platform that runs Docker images as Fly Machines (Firecracker microVMs) in regions around the world with anycast routing, private networking and managed Postgres

    ProprietaryUsage-based
  • PM2

    Production process manager for Node.js and other long-running programs with cluster mode, log handling, restart policies and boot-time startup scripts

    Open sourceOpen source
  • Podman

    Daemonless, rootless container engine with a Docker-compatible CLI, companion Buildah and Skopeo tools and a Podman Desktop GUI

    Open sourceOpen source
  • Slurm Workload Manager

    Open-source cluster resource manager and job scheduler for Linux clusters, allocating nodes, launching jobs and managing queues on many HPC systems

    Open sourceOpen source
  • Traefik Proxy

    Cloud-native reverse proxy and ingress controller that discovers routes from Docker, Kubernetes and other providers and automates TLS through ACME

    MixedOpen source
Compare Kubernetes with its alternatives →

Sources & verification

Primary sources consulted for this record. Claims are tied to the fields each source supports; the automated link checker re-verifies every URL on a schedule.

Confidence
High
Reviewed
Nov 1, 2026
Links
Verified · Nov 2, 2026

Spotted an error?

Broken link, outdated license, wrong category, or a tool that no longer exists? Tell us; reports are reviewed by the editor.

OrbStack

OrbStack (Orbital Labs) · Containers & virtualization

macOS application that runs Docker containers, Linux machines and a local Kubernetes cluster as a lighter replacement for Docker Desktop

ProprietaryFreemiumPaidLocalNov 2, 2026

Trivy

Aqua Security · Security tools

Open-source scanner that finds known vulnerabilities, misconfigurations, exposed secrets and license issues in container images, file systems, git repositories, IaC and Kubernetes clusters

Apache-2.0Open sourceFreeLocalSelf-hostedNov 2, 2026