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

Several vendors combine sovereign-cloud compliance (in-country data residency, isolated tenancy) with turnkey SLURM-as-a-Service on bare-metal GPU fleets:

NorthWind Systems is the most explicit "SLURMaaS" offering.
NorthWind automates bring-up of the underlying Kubernetes cluster and simplifies delivery of per-tenant Slurm clusters on top of it, enabling self-service access and enterprise-grade governance.

NorthWind supports provisioning and lifecycle management across bare metal, virtual machines, Kubernetes, and SLURM environments
, and it's built specifically for
enterprise and sovereign customers with secure, multi-tenant, and air-gapped architectures
. NorthWind also holds an NVIDIA-validated designation for neocloud/sovereign AI factories (northwind.co).

Nscale is a UK/EU sovereign cloud that natively bundles Slurm.
Instances are available as VMs or bare metal nodes, with orchestration via Nscale Kubernetes Service or Slurm clusters, delivered by Nvidia's Slinky — an HPC-grade batch scheduling service that runs Slurm on Kubernetes tuned for large-scale GPU workloads.

Nscale's infrastructure is primarily governed by UK and European laws, operated by local personnel, unlike cloud platforms subject to U.S. extraterritorial laws.

vCluster (with vMetal) targets EU operators building sovereign clouds on bare metal, positioning itself as the layer beneath Kubernetes/Slurm deployments —
delivering the complete infrastructure stack, from bare metal GPU provisioning up through sovereign tenant environments and ready-to-run AI/ML stacks.
Nscale itself uses vCluster underneath, per
Nscale's stack spanning data centers and bare metal to managed Kubernetes, Slurm, and AI services, with vCluster provisioning isolated control planes on a single bare metal underlay cluster.

OpenNebula offers an open-source, self-hostable route:
running Slurm inside OpenNebula combines HPC scheduling with cloud agility via GPU passthrough, and OpenNebula is expanding integration with NVIDIA's Infra Controller for bare-metal GPU server provisioning and lifecycle management from the same platform
— useful for sovereign, self-managed deployments.

Hydra Host is a bare-metal GPU cloud provider explicitly marketed for sovereign deployments:
customers rent GPUs on demand at published hourly rates, reserve capacity, or stand up sovereign-AI deployments, with a public API and self-serve portal.

Sources: northwind.co, nscale.com, vcluster.com, opennebula.io, tooldirectory.ai/tools/hydra-host