A GPU cloud needs tenant boundaries that hold at the control plane and on the hardware. vCluster gives each tenant a virtualized control plane and Private Nodes for its workers, and vCluster Platform adds templates, capacity policy and fleet operations.
Isolating GPU customers from each other means drawing a boundary around both the Kubernetes API and the machines underneath it.
Namespace-only designs share control-plane and node-level components between tenants.
A separate management stack for every tenant increases infrastructure and operational work.
Custom isolation layers take ongoing engineering effort to maintain across cluster, network, and machine lifecycles.
vCluster Platform combines a separate virtualized control plane for every tenant with Private Nodes as the production default. Operators manage tenants, access, templates, capacity, and observability from one platform.
vCluster covers the tenant cluster, worker node, runtime, and optional physical network boundaries needed for GPU cloud tenant isolation.
Private Nodes dedicate worker capacity, networking, and storage to one production tenant cluster at a time.
Every tenant cluster gets its own virtualized control plane, API server, and RBAC boundary on the control plane cluster.
vNode adds a tenant isolation container runtime based on Linux user namespaces and seccomp filters, designed to help prevent container breakout for workloads that need a stronger runtime boundary.
When Netris is configured, separate network environments provide hardware-backed L2 isolation for tenant traffic.
Tenants can create approved cluster products through the built-in UI and CLI or through a custom portal backed by vCluster APIs.
This isn’t a side project. Behind every vCluster deployment is 5+ years of deep K8s engineering, security hardening, and battle-tested infrastructure work at massive scale.
Talk to our team about your stack
Deploy vCluster on your infra in minutes
Go live with a hyperscaler-grade tenant experience in days
Tenant isolation separates each customer's control plane, permissions, workloads, and infrastructure boundary. vCluster provides a virtualized control plane for each tenant, while Private Nodes dedicate worker capacity, networking, and storage to that tenant cluster.
Namespace-only designs share a Kubernetes API server and node-level components. vCluster gives each tenant its own API server and RBAC boundary, then uses Private Nodes as the production default for dedicated worker capacity.
Production tenants do not need to share worker nodes. Private Nodes assign dedicated nodes to one tenant cluster at a time, whether those nodes come from bare metal, VMs, or other Linux machines.
The virtualized control plane comes up in minutes, not hours, and QumulusAI spins up isolated Kubernetes environments in about 1 minute. For bare metal capacity, vMetal provisions a Machine and joins it to the target tenant cluster once the infrastructure driver finishes.
vCluster powers 100K+ GPUs across 50+ GPU clouds and Fortune 500s, and Lintasarta runs 170+ tenant clusters in production. It is validated in NVIDIA's DGX reference architecture.
vCluster Platform supports air-gapped deployments and FIPS features on supported plans. Private Nodes and optional Netris network integration provide deployment controls for regulated or data-residency-sensitive environments.
See how GPU clouds deliver hardware-level tenant isolation with Private Nodes as the production default.
