ai-cloud
Zero Touch Provisioning for AI Cloud Scale
Go from racked GPU servers to a production-ready Kubernetes environment without manual setup. vMetal automates the entire bare metal lifecycle with zero touch provisioning.
Trusted by the fastest-growing AI cloud providers
Problem
Manual Provisioning Slows GPU Revenue
Every hour spent on manual setup is revenue your GPU infrastructure is not generating.
Racks Idle, Revenue Delayed
Manual OS installs and network configs keep GPU servers offline for days or weeks, delaying production workloads and customer onboarding.
Engineering Burned on Setup
Building provisioning pipelines yourself takes 6 to 10 engineers and months of work before a single tenant workload runs.
Human Error Derails Deployments
Manual network configuration and OS installation introduce inconsistencies that cause outages and compliance failures at the worst moments.
Solution
Zero Touch Provisioning From Rack to Kubernetes
vMetal handles PXE boot, OS installation, machine registration, and network automation so your GPU servers reach production without a single manual step. Lintasarta launched Indonesia's leading GPU cloud in 90 days using this stack.
Full Stack Bare Metal to Tenant Clusters
Automate every layer from physical server provisioning to isolated tenant Kubernetes environments running production AI workloads.
Bare Metal Automation
Zero Touch Bare Metal Provisioning
vMetal automates PXE boot, OS installation, machine registration, and full GPU server lifecycle management from rack to production without manual intervention or external dependencies.
- PXE boot and OS install automated
- Machine registration without manual steps
- Full lifecycle from rack to decommission
Kubernetes Distribution
Kubernetes Direct on Bare Metal
vCluster Standalone runs as a single binary directly on bare metal Linux. No k3s, kubeadm, or external Kubernetes dependency required as a base layer before tenant clusters start.
- Single binary on bare metal Linux
- No k3s or kubeadm dependency
- Tenant clusters in seconds after boot
Dynamic Scaling
Auto Provision GPU Nodes on Demand
Auto Nodes acts as bare metal Karpenter, automatically provisioning GPU servers via Terraform when tenants schedule workloads. Physical infrastructure scales to demand without operator intervention.
- Terraform-driven node provisioning
- Triggered by tenant workload scheduling
- No manual scaling operations needed
Network Automation
Automated Per Tenant Network Isolation
Hardware-enforced VLANs, VXLANs, VRFs, ACLs, and DPU policies configure automatically per tenant via Netris integration. Network boundaries are provisioned alongside compute with zero manual configuration.
- VLANs and VXLANs provisioned automatically
- Hardware-enforced per-tenant boundaries
- DPU policy automation via Netris
AI Platform Readiness
Production AI Environments in Minutes
Pre-validated AI stacks including Run:AI, Ray, and Jupyter turn a freshly provisioned Kubernetes cluster into a production AI platform in minutes. No custom integration work required after zero touch provisioning completes.
- Run:AI, Ray, Jupyter pre-validated
- Cluster to AI platform in minutes
- Certified with tenant isolation
Why vCluster
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
50+
GPU Clouds & F500s
<45
Days to Launch
30K
GitHub Stars
Get Started in 3 Steps
1
Schedule a Demo
Talk to our team about your stack
2
Deploy vCluster
Deploy vCluster on your infra in minutes
3
Onboard Your Tenants
Go live with a hyperscaler-grade tenant experience in days
FAQs
What is zero touch provisioning for GPU servers?
Zero touch provisioning automates the entire process of bringing a GPU server from physical rack installation to a production-ready Kubernetes environment. This includes PXE boot, OS installation, machine registration, and network configuration — all without manual operator intervention. vMetal implements zero touch provisioning for bare metal GPU infrastructure, handling each stage of the server lifecycle automatically so engineering teams can focus on delivering services rather than configuring hardware.
Does zero touch provisioning work without an existing Kubernetes cluster?
Yes. vCluster Standalone runs as a single binary directly on bare metal Linux with no external Kubernetes dependency. You do not need k3s, kubeadm, or any other base layer pre-installed. This means zero touch provisioning can take a freshly imaged server straight to a running Kubernetes distribution and tenant clusters without intermediate setup steps.
How does network configuration work in zero touch provisioning?
vMetal integrates with Netris to automate network provisioning alongside compute. VLANs, VXLANs, VRFs, ACLs, and DPU policies are configured automatically per tenant as part of the provisioning pipeline. Hardware-enforced network isolation is established without manual switch configuration or separate network engineering work.
Can zero touch provisioning scale GPU nodes automatically as demand grows?
Auto Nodes extends zero touch provisioning to dynamic scaling. It functions as bare metal Karpenter, automatically provisioning additional GPU servers via Terraform when tenants schedule workloads that require more capacity. Physical GPU nodes are added to the cluster without operator involvement, so your infrastructure scales to meet demand without a provisioning backlog.
How quickly can a GPU cloud launch using zero touch provisioning?
Lintasarta launched Indonesia's leading GPU cloud in 90 days using the vMetal and vCluster stack, deploying 170 or more tenant clusters. Boost Run launched a managed Kubernetes offering in less than 45 days without adding platform engineering headcount. Zero touch provisioning removes the manual bottlenecks that typically extend GPU infrastructure timelines by months.
Is zero touch provisioning supported in air-gapped or compliance environments?
vCluster supports air-gapped and FIPS deployment environments. The provisioning stack can operate in disconnected data centers where servers cannot reach external package repositories or registries. This makes zero touch provisioning viable for regulated industries and government data centers that require network-isolated infrastructure without sacrificing automation.
Automate Your GPU Server Provisioning Today
See how vMetal delivers zero touch provisioning from rack to production Kubernetes.
