Using Kubernetes#
SkyPilot tasks can be run on your private on-prem or cloud Kubernetes clusters. The Kubernetes cluster gets added to the list of βcloudsβ in SkyPilot and SkyPilot tasks can be submitted to your Kubernetes cluster just like any other cloud provider.
Why use SkyPilot on Kubernetes?#
No complex kubernetes manifests - write a simple SkyPilot YAML and run with one command sky launch
.
SSH access to pods, VSCode integration, job management, autodown idle pods and more.
Kubernetes cluster is full? SkyPilot seamlessly gets resources on the cloud to get your job running sooner.
Train and serve Llama-3, Mixtral, and more on your Kubernetes with ready-to-use recipes from the AI gallery.
Scale beyond your Kubernetes cluster to capacity on across clouds and regions without manual intervention.
SkyPilot can run with your custom pod scheduler and automatically terminate idle pods to free up resources for other users.
Works with your existing observability and monitoring tools, such as the Kubernetes Dashboard.
Reduce operational overhead by letting your teams provision their own resources, while you retain control over the Kubernetes cluster.
Table of Contents#
Features and Roadmap#
Kubernetes support is under active development. Some features are in progress and will be released soon:
CPU and GPU Tasks - β Available
Auto-down - β Available
Storage mounting - β Available on x86_64 clusters
Multi-node tasks - β Available
Custom images - β Available
Opening ports and exposing services - β Available
Multiple Kubernetes Clusters - π§ In progress