Skip to content

Multi-Tenancy#

conceptstub

Sharing one GPU cluster between teams safely and fairly.

  • Concepts

    How it works: the mechanism, the moving parts, and where it sits in the stack.

  • Design

    Trade-offs, decision records, and when to choose which option.

  • Implementation

    Manifests, Helm values, and scripts. Every script states whether it is read-only or changes the cluster.

  • Testing

    How to prove it works: test procedures, expected output, and performance baselines.

  • Troubleshooting

    Failure modes, symptoms, and the diagnostic path from symptom to root cause.

Planned content

  • Tenant model: namespace-per-tenant with labels and RBAC
  • Quotas: ResourceQuota for GPUs, Kueue ClusterQueues and cohorts
  • Network isolation: Calico policy on the primary network, per-tenant secondary networks
  • GPU isolation tiers: dedicated nodes, MIG, HAMi
  • Guardrails: OPA Gatekeeper constraints