Canonical has released MicroK8s - a fast and efficient upstream Kubernetes delivered as a single snap package that installs on 42 flavors of Linux. With a small disk and memory footprint, MicroK8s provides an efficient way to deploy Kubernetes in seconds, whether on the desktop, the server, an edge cloud, or IoT device.
MicroK8s is available through the Snap Store and can be deployed easily:
sudo snap install microk8s --classicThe benefits of providing MicroK8s as a snap include automatic updates and well-defined security capabilities. Automatic updates ensure developers are always working from the latest upstream Kubernetes with binaries delivered directly from the source and configured in seconds. Running the latest upstream version also means that MicroK8s benefits from Kubernetes’ built-in security features. MicroK8s installs fully conformant Kubernetes version 1.13 and comes with exciting new features.
To further accelerate Kubernetes adoption and simplify common developer scenarios, MicroK8s includes a growing number of add-on services, including a container registry, storage pass-through and native GPGPU enablement – all of which are enabled with a single command. For data scientists and machine learning engineers, the GPGPU enablement simplifies adding hardware acceleration to their machine learning workflows.
Key use cases that MicroK8s facilitates include:
-Embedded, auto-updating Kubernetes to power IoT applications
-Setup a CI/CD pipeline locally in a few steps
-Quickly setup disposable Kubernetes as part of your production CI/CD pipeline
-Deploy single-node applications in a scaled-up server
-Create a local OCI-compliant container registry to cache commonly used containers
-Experiment with CNCF Trail Map projects easily and quickly
-Accelerate machine learning modeling and training with GPU support
-Kubeflow Deployments – the open-source ML toolkit for Kubernetes.
MicroK8s is free to install from the Snap Store by clicking here. Additionally, for developers wishing to extend MicroK8s, join the community on GitHub!
Comentários