nebius/soperator

Run Slurm in Kubernetes

55
/ 100
Established

This project lets machine learning engineers and researchers run their demanding computational workloads, like training large models or performing high-performance simulations, on a Slurm cluster that operates within a Kubernetes environment. It takes your existing Slurm scripts and configurations and runs them, providing the robust job scheduling and resource management benefits of Slurm alongside the infrastructure reliability and scalability of Kubernetes. This is ideal for ML engineers, data scientists, and HPC users who need to manage complex, distributed computational jobs.

368 stars.

Use this if you need to run high-performance computing tasks or distributed machine learning model training jobs and want to combine Slurm's specialized job scheduling with Kubernetes's operational benefits like auto-scaling and self-healing.

Not ideal if your computational tasks are simple and don't require the advanced scheduling or resource isolation of Slurm, or if you need to run Slurm clusters with mixed GPU/CPU nodes or multiple partitions.

machine-learning-operations high-performance-computing model-training cluster-management research-computing
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 19 / 25

How are scores calculated?

Stars

368

Forks

54

Language

Go

License

Apache-2.0

Last pushed

Mar 13, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/mlops/nebius/soperator"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.