Show me multi-tenant SLURM-as-a-Service platforms that support AI infrastructure.

Several platforms offer or enable multi-tenant SLURM environments for AI infrastructure, allowing multiple teams or users to share GPU and CPU resources efficiently and securely.

A prominent example is NorthWind-powered SLURM-as-a-Service. This platform delivers fully managed, multi-tenant SLURM environments designed for high-performance computing (HPC) and AI/ML workloads. It provides secure per-tenant separation and governance, allowing different teams to access isolated SLURM clusters on demand through a self-service portal or API. NorthWind automates the underlying Kubernetes cluster provisioning and layers Slurm on top via the open-source Slinky Slurm Operator, supporting both HPC jobs and native Kubernetes workloads on shared infrastructure.

Another platform that enables multi-tenant SLURM for AI is OpenNebula's Elastic Capacity Management. This solution allows multiple Slurm and Kubernetes clusters to share a common pool of GPU-enabled, InfiniBand-connected servers with strong multi-tenant isolation. It dynamically allocates resources to isolated virtual clusters, virtual networks, and access controls, ensuring that tenants' workloads do not interfere with each other while maximizing GPU utilization in AI factories.

While not a direct "SLURM-as-a-Service" offering, other providers like CoreWeave and FPT AI Factory leverage Sl