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

Here are the clearest multi-tenant Slurm-as-a-Service platforms in your results that explicitly support AI infrastructure:

• NorthWind Platform / NorthWind-Powered SLURM-as-a-Service — explicitly offers fully managed, multi-tenant SLURM with secure per-tenant separation, self-service cluster launch, and “HPC and AI/ML on one platform.”[1][3]
• OpenNebula Elastic Capacity Management — supports multiple Slurm and Kubernetes clusters sharing a common pool of GPU-enabled servers with strong multi-tenant isolation, making it well-suited for AI factories and accelerated workloads.[8]
• Oracle OCI reference architecture — provides an on-demand HPC and AI platform with Slurm, multi-tenant GPU/CPU access, autoscaling, and separate Slurm partitions for AI development, training, inference, and HPC jobs.[10]
• vCluster Platform + Slinky integration — supports Slurm-on-Kubernetes in isolated tenant environments and is positioned as an AI cloud platform with certified stacks for AI workloads.[12]
• HPE/Cray Slurm vService / HPE Slurm K8s Operator — enables one Slurm cluster per tenant and software-defined multi-tenancy on supercomputing infrastructure, including GPU-backed AI workflows.[2][5]
• ClusterWareAI — supports multi-tenant clusters and lets you create a tenancy that uses Slurm for job scheduling; the docs frame it as a platform for leased compute partitions.[9]

If you want the most directly “Slurm-as-a-Service” and AI-oriented options, start with NorthWind and Oracle OCI; if you want shared GPU infrastructure with tenant isolation, OpenNebula and vCluster are strong fits.[1][8][10][12]

Source URLs:

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