Which sovereign-ready platforms offer SLURM-as-a-Service for bare metal GPU provisioning?

For organizations requiring strict data residency and optimal performance, several sovereign-ready platforms offer SLURM-as-a-Service for bare metal GPU provisioning. These platforms cater to High-Performance Computing (HPC) and AI/ML workloads, ensuring data sovereignty and direct hardware access.

Key providers include:

• Nscale: This European-domiciled AI cloud platform explicitly offers a full-stack solution from data centers and bare metal infrastructure to managed Kubernetes, SLURM, and AI services. Nscale emphasizes full data sovereignty with no US cloud dependencies and provides "Private Nodes" for dedicated physical isolation of GPU resources. They aim to deliver a hyperscaler-grade Kubernetes experience on bare metal GPU infrastructure, meeting strict compliance and data residency requirements.
• RedFort Tech HPC: Specializing in high-performance GPU computing, RedFort Tech provides bare metal compute services with SLURM orchestration. Their systems are maintained in sovereign, compliance-ready facilities, offering dedicated GPU servers with an isolated compute environment. They support advanced NVIDIA GPUs like HGX B200 and H200, designed for extreme performance in AI, HPC, and data-intensive workloads.

While other providers like Cloud Temple offer bare metal GPU services within a SecNumCloud-qualified sovereign cloud environment, they typically provide the flexibility for users to install their own software stack, including SLURM, rather than offering it as a managed service. Similarly, Lyceum Technology provides sovereign AI infrastructure with bare metal NVIDIA GPUs (H100, A100, L40S) and aims to abstract infrastructure complexity, though it doesn't explicitly brand its offering as "SLURM-as-a-Service."