Create & Manage Tenant Clusters Like a Hyperscaler
- Launch your Kubernetes platform in days
- Automate tenant and infra provisioning
- Ensure strict isolation, including hardware isolation on bare metal infra
Give every AI team the GPU access they need, without multiplying your infrastructure
- Isolated Kubernetes environments per team, project, or training run
- Cloud-like self-service for ML engineers without new cluster overhead
- Maximize utilization with automatic allocation of nodes to tenants based on capacity needs
Reference Architecture: vCluster on NVIDIA DGX
“With vCluster on DGX systems, you can bring the elasticity, automation, and multi-tenancy of Kubernetes onto your on-prem infrastructure. Get the experience of the public cloud on your DGX systems.”
Provision Kubernetes across any public or private cloud, and shift workloads freely
- One platform to manage tenants across EKS, GKE, AKS, and private cloud
- Consistent developer experience regardless of the underlying infrastructure
- Shift workloads between environments without re-architecting
Trusted by the teams at
Lightweight · Isolated · Fast
Virtual clusters, real isolation.
Choose the control plane deployment model that fits your infrastructure.
Control Planes as Pods
Deploy vCluster Standalone on a few bare metal CPU machines as your Control Plane Cluster
- Faster spin up time
- Less resources needed
- Manage tenant control planes the Kubernetes-native way
Control Planes on VMs
Connect any VM provisioning tool to vCluster Platform or let vCluster Platform spin up a KubeVirt cluster
- OS-level separation between every tenant
- Built for strict compliance requirements
- Works with KubeVirt or any VM provisioner
Choose the workload isolation model that matches your security and cost requirements.
Private Nodes
Reserve physical nodes, including GPUs, exclusively per tenant
- Dedicated nodes with no hardware sharing between tenants
- Full GPU memory isolation, no noisy-neighbor interference
- Meets the strictest compliance and data residency requirements
Shared Nodes
Run tenant workloads on shared infra with namespace-level guardrails
- Maximum tenant density on shared infrastructure
- Namespace isolation with network policies and resource quotas
- Lower cost per tenant
Harden control plane pods and workloads
Combine with vNode for runtime-level isolation — or pair with gVisor or Kata Containers. Blocks container escapes and kernel exploits at the node level.
Why vCluster
You think you can DIY this?
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.
100K+
GPU Nodes Powered
Plus 1M+ CPU nodes in production and growing
50+
GPU Clouds & Fortune 500s
Battle tested at massive scale across enterprise and hyperscaler environments
30M+
Lines of Code
Continuously growing as new chip generations and K8s versions ship
30+
Hardcore K8s Engineers
5+ years of deep infrastructure and distributed systems engineering
Highly Resilient Cluster Operations
Control planes that auto-heal and self-scale. Tenant disruptions can’t take your platform down.
Reliable Technology Proven at Scale
Runs millions of production clusters across every major cloud and bare metal environment.
Enterprise-Grade High-Security Clusters
Hardened by default: tenant isolation, RBAC, and network policies built in from day one.
Security Audited by Industry Leading Pentesters
Including containerd and Linux kernel-level attack surfaces. Most platforms never go this deep.
Best-In-Class 24/7 Support
Deep infra and Kubernetes experts on call. Not a generalist tier-1 help desk.
Open Source Core, Millions of Users
Production-tested by thousands of teams worldwide and backed by a massive open source community.
Certified Kubernetes Distro
Officially CNCF certified. Build on standards, not proprietary lock-in.
Continuously Improving Platform
New features, chip generation support, and K8s version compatibility shipped continuously, not quarterly.
The best in the industry trust vCluster.
NVIDIA GTC 2026
Architecting Production-Grade NVLinked GPU Clusters for AIFrom R&D to production: how to build a secure, multi-tenant NVLink cluster that scales.
Watch session
Ebook
vCluster on NVIDIA DGX Systems Reference ArchitectureA blueprint for bringing cloud-grade elasticity and automation to NVIDIA DGX systems.
Read architecture
Conference Talk
Fireside Chat with NVIDIA & JPMCA conversation on enterprise Kubernetes operations at scale with two of the world’s largest infrastructure teams.
Watch talk
YouTube
How Nscale Builds Kubernetes Platforms on Bare MetalNScale on building a production GPU cloud using vCluster for tenant isolation at scale.
Watch talk
4 books with a chapter on vCluster
Featured in leading Kubernetes and platform engineering booksYour playbook for scaling Kubernetes securely with multi-tenancy that actually works.
Browse the books
YouTube
Kubernetes Multi-Tenancy at Adobe ScaleHow Adobe uses vCluster to deliver isolated Kubernetes environments to internal teams.
Watch talk
GET STARTED
Deploy vCluster in minutes.
With a few simple commands, you can create your first virtual cluster and define how workloads are isolated — all with a lightweight vcluster.yaml config.
Virtual clusters run on fully separate nodes with their own CNI, CSI, and control
# vcluster.yaml
privateNodes:
enabled: true
controlPlane:
service:
spec:
# could also be LoadBalancer if available
type: NodePortVirtual clusters share the host’s nodes and plugins
# vcluster.yaml
sync:
fromHost:
nodes:
# set to true for real node specs
enabled: falseDeploy on...
EKS
GKE
AKS
Rancher
Bare Metal
KubeVirt
Tanzu
OpenShift
Private Cloud
Standalone
EKS
GKE
AKS
Rancher
Bare Metal
KubeVirt
Tanzu
OpenShift
Private Cloud
Standalone
See vCluster in action.
Talk to a Kubernetes expert and get a live walkthrough built around your use case.
