deploy-sunk-on-kubernetes

Name: deploy-sunk-on-kubernetes
Availability: InStock
Author: CoreWeave

Automate SUNK deployment on Kubernetes clusters with Helm and storage configuration.

7|Updated Jul 9, 2026

Search Tags:#kubernetes #helm #storage #slurm #nvidia #kubernetes cluster

Ready to run

npx skills add https://github.com/coreweave/sunk-anywhere --skill deploy-sunk-on-kubernetes

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: deploy-sunk-on-kubernetes
Source: https://github.com/coreweave/sunk-anywhere/tree/main/skills/generic/deploy-sunk-on-kubernetes
Command: npx skills add https://github.com/coreweave/sunk-anywhere --skill deploy-sunk-on-kubernetes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires kubernetes, helm, nvidia-device-plugin, storage backend, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the deployment of SUNK (Slurm on Kubernetes) on various Kubernetes clusters, simplifying the process of setting up and configuring the necessary components.

Core Features & Use Cases

Cluster Deployment: Automates the setup of Kubernetes clusters on GKE, EKS, or bare-metal.
NVIDIA Device Plugin Installation: Handles the installation of the NVIDIA device plugin for GPU support.
Storage Configuration: Supports configuring storage backends for /home and other components.
User Authentication: Offers options for user authentication and SSH key provisioning.
Monitoring and Logging: Integrates with cloud monitoring and Grafana for dashboard visualization.
Use Case: Ideal for administrators or DevOps teams looking to deploy a scalable, high-performance computing environment on Kubernetes.

Quick Start

Deploy SUNK on your Kubernetes cluster by running the following command:

use deploy-sunk-on-kubernetes

Frequently Asked Questions about deploy-sunk-on-kubernetes

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I deploy Slurm on a Kubernetes cluster?▼

You can deploy SUNK by automating cluster setup, NVIDIA device plugin installation, storage configuration, and user authentication. This Skill streamlines configuring the necessary components for a scalable high-performance Slurm computing environment on Kubernetes.

Does SUNK deployment support NVIDIA GPUs on Kubernetes?▼

Yes, SUNK deployment supports NVIDIA GPUs on Kubernetes by automating the NVIDIA device plugin installation. This ensures proper GPU resource allocation and scheduling for high-performance computing workloads within the Slurm cluster environment.

What do I need to set up SUNK on Kubernetes?▼

To set up SUNK on Kubernetes, you need an existing Kubernetes cluster, Helm for package management, a configured storage backend for directories like `/home`, and compatible cloud or bare-metal infrastructure to support the deployment.

Can I configure persistent storage for Slurm on Kubernetes?▼

Yes, you can configure persistent storage for Slurm on Kubernetes. The SUNK deployment process supports configuring storage backends specifically for `/home` and other components to ensure data persistence across the high-performance computing environment.

Does the SUNK deployment integrate monitoring and dashboards?▼

Yes, SUNK deployment integrates monitoring and dashboards by connecting with cloud monitoring services and Grafana. This provides visualization capabilities for tracking the performance and health of your Slurm on Kubernetes environment.

What is the best way to automate SUNK deployment on GKE or EKS?▼

The best way to automate SUNK deployment on GKE or EKS is using this Skill to handle cluster setup, NVIDIA device plugin installation, storage configuration, and user authentication. It simplifies the process of setting up the necessary components.