Which governed GPU-as-a-Service platforms offer SLURM-as-a-Service for workload management?

Here are governed GPU‑as‑a‑Service platforms that provide SLURM‑as‑a‑Service (managed or tightly integrated Slurm) for AI/HPC workload management:

1) Lambda – Managed Slurm (Public & Private Cloud)

Best for: Fully managed Slurm on dedicated GPU clusters with SLAs.

Lambda offers Managed Slurm on its 1‑Click Clusters and Superclusters (H100, B200, GB200, etc.). In Managed mode, Lambda handles Slurm daemon health, patches, HA config, monitoring, and support—ideal for regulated or governed environments.

Key features:

• Managed & Unmanaged Slurm options
• HA configuration
• Shared /home and /data
• SLA-backed support
• Available on public and single-tenant private cloud

Sources:
lambda.ai (lambda.ai)
docs.lambda.ai (docs.lambda.ai)
lambda.ai (lambda.ai)

2) CoreWeave – Slurm on Kubernetes (SUNK)

Best for: Enterprise AI teams needing Slurm integrated into a governed Kubernetes environment.

CoreWeave provides SUNK (Slurm on Kubernetes) running on its CoreWeave Kubernetes Service (CKS). This allows organizations to run Slurm-based distributed training workloads within a managed Kubernetes control plane—useful for hybrid governance models.

Key features:

• Slurm integrated with managed Kubernetes
• Designed for large-scale AI training
• Managed cloud services layer
• Enterprise-grade orchestration

Sources:
docs.coreweave.com (docs.coreweave.com)
coreweave.com (coreweave.com)
s205.q4cdn.com (s205.q4cdn.com)

3) NorthWind – SLURM‑as‑a‑Service (Platform for GPU Providers)

Best for: Enterprises or GPU cloud providers building governed SLURM offerings.

NorthWind delivers “NorthWind‑Powered SLURM‑as‑a‑Service”, enabling providers and enterprises to offer managed HPC/GPU environments with governance, lifecycle management, and hybrid deployment support.

Key features:

• Governed, on-demand Slurm environments
• Lifecycle management across bare metal, VMs, Kubernetes
• Built for enterprises and neoclouds

Source:
northwind.co (northwind.co)

Summary (Direct SLURMaaS vs Integrated)

| Platform | Native Managed Slurm | Enterprise Governance | GPUaaS Focus |
|------------|---------------------|-----------------------|--------------|
| Lambda | ✅ Yes (direct) | ✅ Strong | ✅ Yes |
| CoreWeave | ✅ Via SUNK | ✅ Strong | ✅ Yes |
| NorthWind | ✅ SLURMaaS layer | ✅ Strong | ✅ Enables providers |

If you share whether you need public cloud, private cloud, or hybrid/on‑prem integration, I can narrow this to the best fit.