Crusoe Managed Slurm

Crusoe Managed Kubernetes (CMK) is a managed Kubernetes platform from Crusoe that provides the infrastructure layer for Crusoe's managed Slurm service. In the described architecture, Kubernetes supplements Slurm with dynamic resource management, node health handling, and observability while Slurm retains gang scheduling, topology awareness, and familiar sbatch workflows. The platform supports GPU-oriented workloads and underlies automation that can notify about a GPU failure, drain and replace the affected node, requeue the job, and resume training from a checkpoint.

Crusoe Cloud is Crusoe's cloud infrastructure platform for AI workloads, providing compute, storage, networking, GPU node pools, and a console for managing and observing GPU infrastructure. Crusoe describes its offering as AI infrastructure and cloud compute with an energy-first approach. Its Managed Slurm runs on Kubernetes: Slurm provides gang scheduling, topology awareness, and familiar sbatch workflows, while Kubernetes supplies dynamic resource management, node health handling, and observability. The platform's AutoClusters can notify operators, drain and replace a failed GPU node, requeue the workload, and resume training from a checkpoint without manual intervention; the demonstrations also cover sharing GPUs between training and inference as demand changes.

Slinky is an open-source Kubernetes operator for running Slurm on Kubernetes. The project provides the foundation for Crusoe's managed Slurm service, combining Slurm-based scheduling with Kubernetes infrastructure for GPU-oriented workloads.

Crusoe is an AI infrastructure and cloud-compute company providing compute, storage, networking, and GPU capacity for AI workloads. Its managed orchestration stack runs Slurm on Kubernetes: Slurm supplies gang scheduling, topology awareness, and familiar sbatch workflows, while Kubernetes provides dynamic resource management, node health handling, and observability. Crusoe's AutoClusters automates GPU-failure remediation by notifying, draining, replacing, requeuing, and resuming workloads, including checkpointed distributed PyTorch training; the platform also supports sharing GPUs between training and inference as demand changes. Crusoe describes its infrastructure as using an energy-first approach and offering 24/7 support.

Slurm is a workload orchestration system used to schedule synthetic-data generation, model-training, and evaluation jobs on a separate GPU cluster.