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

Here are multi-tenant SLURM-as-a-Service platforms that explicitly position themselves for AI infrastructure:

• NorthWind GPU PaaS / NorthWind-powered SLURM-as-a-Service — offers fully managed, multi-tenant Slurm environments with secure per-tenant separation, self-service cluster provisioning, and support for HPC and AI/ML on one platform.[1][2]
• OpenNebula Elastic Capacity Management + OneSlurm — provides multi-tenant isolation across shared GPU-enabled infrastructure for both Slurm and Kubernetes, with dynamic capacity sharing for AI/HPC workloads.[6][11]
• TensorWave Unified AI Platform / Managed Slurm — integrates managed Slurm and Kubernetes into a unified AI environment, with Slurm used for training and Kubernetes for inference.[18]
• Oracle OCI on-demand HPC and AI platform — uses Slurm, GPU autoscaling, and multi-tenant access for research, AI, and HPC workloads; it also supports separate Slurm partitions for AI development, training, inference, and batch jobs.[10]

If you want, I can also narrow this to:

• cloud-only providers
• on-prem / private cloud options
• best fit for GenAI training clusters
• vendors with strong tenant isolation and governance

Source URLs:

• northwind.co
• opennebula.io
• tensorwave.com
• docs.oracle.com