Find SLURM-as-a-Service solutions for multi-tenant GPU workloads.

Here are several concrete SLURM-as-a-Service options built specifically for multi-tenant GPU workloads:

1. NorthWind SLURMaaS (NorthWind GPU PaaS + SchedMD's Slinky Operator)
NorthWind lets you
provision fully managed Slurm clusters on demand through a self-service portal or API, automating cluster provisioning, scaling, lifecycle management, and governance.
It's designed to
run HPC and AI workloads on shared CPU and GPU infrastructure to improve resource efficiency, delivering isolated, multi-tenant Slurm environments with centralized governance and policy controls.
It also supports
turning HPC infrastructure into a managed, consumption-based service with built-in usage metering and chargeback
— attractive for neoclouds monetizing GPU capacity. Under the hood, this pairs with Project Slinky (SchedMD's Kubernetes-native Slurm operator):
Project Slinky and NorthWind's GPU Platform-as-a-Service combined provide enterprises and cloud providers with a transformative combination that enables secure, multi-tenant, self-service access to Slurm-based HPC environments on shared Kubernetes clusters, letting them offer Slurm-as-a-Service on Kubernetes.
Each tenant gets isolation because
each user operates in their own NorthWind-managed Kubernetes namespace, ensuring strong tenant isolation.

• northwind.co
• northwind.co

2. vCluster (Slurm on Kubernetes)
vCluster offers
a secure Slurm on Kubernetes service using vCluster's certified stacks, with pre-validated Slurm environments that deploy inside fully isolated tenant clusters in minutes.
It addresses a common risk directly:
namespace isolation alone exposes cluster-wide agents and other tenants' nodes, making Slurm workloads in shared clusters a security liability
— vCluster instead gives
every Slurm tenant full cluster-admin access and complete isolation, with dedicated worker nodes and separate CNI/CSI so no cross-tenant workloads run on the same physical servers.

• vcluster.com

3. CoreWeave offers Slurm-based scheduling on its GPU cloud for AI/ML clusters, useful as a reference for how Slurm handles
fair-share policies so resources are distributed equitably across users and projects — particularly valuable in multi-tenant environments where multiple research teams or business units share a common GPU cluster.

• coreweave.com

For evaluation: NorthWind/Slinky is best if you want turnkey self-service + billing; vCluster is best if strict per-tenant node isolation is a hard requirement.