Introduction
An overview of the HPC Dashboard for SLURM clusters
Welcome to the HPC Dashboard! This powerful Next.js application provides comprehensive monitoring and management tools for your SLURM-based High-Performance Computing cluster. With real-time visualization, detailed job tracking, and AI-powered assistance, the HPC Dashboard transforms how you interact with your computing resources.
Open Source Philosophy
HPC Dashboard is proudly open-source and designed for the HPC community! 🎉 We believe in creating accessible, powerful tools that make cluster management more intuitive.
Contribute
Interested in helping improve the HPC Dashboard? Check out our GitHub repository to get started! From feature suggestions to bug fixes, all contributions are welcome.
Project Overview
HPC Dashboard is a comprehensive toolkit designed for modern cluster administrators and users. Key features include:
- Real-time Monitoring: Track cluster utilization, job status, and node health in real-time
- Interactive Visualizations: Intuitive, graphical representation of complex cluster data
- AI-powered Chat: Get answers about SLURM commands, job details, and best practices
- Embeddings Support: Search documentation and receive contextually relevant answers
Key Features
| Feature | Description |
|---|---|
| Node Monitoring | Real-time visualization of CPU, memory, and GPU utilization |
| Job Tracking | Detailed job information with status updates and history |
| SLURM Integration | Direct interfacing with SLURM's REST API |
| OpenAI Chat Assistant | AI-powered support for SLURM queries and cluster information |
| Prometheus Metrics | Integration with Prometheus for advanced performance metrics |
| Documentation Search | Embeddings-based search for finding relevant information |
| Module System Integration | Display and search available LMOD/Environment modules |
Technology Stack
HPC Dashboard leverages cutting-edge technologies to ensure performance and flexibility:
- Next.js - Server components and app router architecture for dynamic content
- TypeScript - Type-safe coding for reliability and maintainability
- Tailwind CSS - Utility-first styling for responsive, clean interfaces
- Shadcn/UI - Elegant, accessible components built on Radix UI
- AI SDK - OpenAI integration for intelligent assistant capabilities
- Drizzle ORM - Type-safe database operations for embedding storage
- PostgreSQL Vector DB - Efficient storage and retrieval of document embeddings
- Recharts - Interactive data visualization components
Directory Structure
Deployment
The HPC Dashboard is best deployed locally using PM2
For detailed deployment instructions, check out our Deployment Guide.
Support and Community
For support with the HPC Dashboard, you can:
- Open an issue on our GitHub repository
- Join our community on Discord
We're dedicated to making HPC management more accessible and intuitive through this dashboard, and we welcome feedback and contributions from the community!
Made with ❤️ for HPC