microk8s
MicroK8s is a small, fast, single-package Kubernetes for datacenters and the edge.
A lightweight Kubernetes distribution from Canonical that installs with a single Snap command on laptops, edge devices, and servers, with optional add-ons for monitoring, networking, and GPU workloads.
MicroK8s is a lightweight version of Kubernetes, the system that large companies use to run and manage software applications across many servers. Kubernetes is normally complex to install and operate, but MicroK8s packages everything into a single installable unit that runs on a laptop, a developer workstation, a small IoT device, or an edge server. It is made by Canonical, the company behind the Ubuntu Linux distribution.
Installation is a single command using the Snap package manager, which is the standard way to install software on Ubuntu and many other Linux distributions. Once installed, MicroK8s provides a built-in version of kubectl, which is the command-line tool used to interact with Kubernetes. It also creates a user group so that you can run commands without needing administrator privileges each time.
Out of the box, MicroK8s is a minimal Kubernetes cluster. Optional components can be turned on one at a time using the enable command. The available add-ons include a web dashboard for viewing what is running, DNS for internal name resolution, monitoring tools (Prometheus and Grafana), networking tools (Istio and Linkerd), and GPU support for machine learning workloads. A full list of available and currently enabled add-ons is visible with the status command.
The project tracks Kubernetes releases closely, publishing beta, release candidate, and final builds on the same day as the upstream Kubernetes project. You can stay on the latest version or pin to any specific version from 1.10 onward. It is certified as a conformant Kubernetes implementation, meaning applications built for major cloud-managed Kubernetes services like Amazon EKS or Google GKE should also work on MicroK8s.
It is free and open source, with official documentation maintained separately at microk8s.io.
Where it fits
- Run a full Kubernetes cluster on your laptop for local development without cloud costs.
- Deploy Kubernetes on a Raspberry Pi or IoT edge device with a single install command.
- Test Kubernetes add-ons like Prometheus, Istio, or Linkerd locally before deploying to a cloud cluster.
- Set up a GPU-enabled Kubernetes node for running machine learning workloads locally.