I’d like to share Slurm-web, an open source web interface to monitor Slurm HPC & AI clusters in real time.

It provides a read-only overview of job queues and node status with metrics and advanced visualizations – useful to sysadmins, managers and end-users – without needing to SSH in or run command-line tools.

Slurm is the world's leading workload manager and the de facto standard for HPC and AI clusters. It has a powerful command-line interface, but it's complex to use. Many HPC centers deploy advanced stacks to create custom dashboards with exporters, Prometheus, InfluxDB, Grafana, which are complex and time-consuming to design and maintain.

Slurm-web offers clear and accessible views of the cluster – with minimal setup and overhead.

And it is released as free software under the GNU GPL v3!

Key features:

  - Dashboard with interactive charts of resources and jobs status
  - Instant jobs filtering and sorting
  - Live jobs status update with colored badges to visualize at a glance
  - GPU resources utilization monitoring
  - Advanced visualization of node status with racking topology
  - Intuitive visualization of QOS and advanced reservations
  - Dark mode support
  - Multi-clusters support
  - LDAP authentication (including Active Directory support)
  - Advanced RBAC permissions management
  - Transparent caching
  - Integration with Prometheus to collect and chart timeseries metrics of Slurm

  - Backend: Python (Flask), uses Slurm REST API to consume data from scheduler
  - Frontend: Vue.js and Tailwind CSS

The roadmap is full of features such as reporting and job submissions. In the meantime, I would be more than happy to get your feedback on this software!

Comments URL: https://news.ycombinator.com/item?id=44170750

Points: 16

# Comments: 0

It's fun, with good mix of interesting technologies and it seems to solve a real problem for many users in HPC community!

https://slurm-web.com/