Articles

vCluster Platform v4.12 and vCluster v0.37 are supported together for 18 months, with fixes and security updates through April 15, 2028.

Operators define the environment once and hand it to every tenant the same way, so a new managed service is something they ship rather than staff.

vCluster Labs today announced the general availability of Stacks in vCluster Platform v4.12, so an operator can deploy the same environment to every tenant.

Managed AI infrastructure earns more than raw capacity, but only if you can staff every environment you sell. Certified Stacks remove that constraint, starting with NVIDIA Run:ai.

Managed AI infrastructure earns more than raw capacity, but only if you can staff every environment you sell. Certified Stacks remove that constraint, starting with NVIDIA Run:ai.

Define the NVIDIA Run:ai setup once, with its dependencies, readiness gates, and outputs, then give every new team a working tenant cluster from the same template.

Installing NVIDIA Run:ai means wiring up registry credentials, ingress, GPU components, the control plane, and cluster registration in the right order. This walkthrough uses vCluster Platform 4.12 Stacks to turn that sequence into a reusable template, create a tenant cluster from it, and run your first project.

Give each team its own Kubernetes cluster on a single DGX Spark, and let their workloads call two resident Minima models

Give each team its own Kubernetes cluster on a single DGX Spark, and let their workloads call two resident Minima models. One GPU system, two useful models, and more than one team that wants to use them. That is the shape of most local AI i

Build a complete GitOps pipeline with GitHub Actions and Argo CD on a multi-node Kubernetes cluster running in Docker.

GitOps works best when you can break the whole loop and rebuild it. This walkthrough goes end to end: a multi-node Kubernetes cluster in Docker via vCluster, GitHub Actions building and tagging images, and Argo CD pulling each change.

What NVIDIA's new GPU burn-in tool can show about a tenant cluster before handoff, tested on two A100 Private Nodes.

A Kubernetes node can report Ready while the system behind it is nowhere near ready for distributed AI. The kubelet is healthy. The API server can reach it.

This is a simple, step-by-step guide to vCluster logging and audit logs: control plane logs, JSON logs, debug logs, and Kubernetes audit logs!

This is a simple, step-by-step guide to vCluster logging and audit logs: control plane logs, JSON logs, debug logs, and Kubernetes audit logs. Every step includes the exact commands and example output so you know what to expect

Observability becomes harder when the question shifts from "Is this cluster healthy?" to "Is the fleet healthy?" Run a complete system with vCluster Fleet Observability

Observability becomes harder when the question shifts from "Is this cluster healthy?" to "Is the fleet healthy?" Running a complete monitoring stack in every tenant cluster creates duplicated infrastructure, while scraping every cluster fro

Make AuthorizationPolicy work in tenant clusters with custom resource sync

Make AuthorizationPolicy work in tenant clusters with custom resource sync vCluster makes it easy to give each team a “real” Kubernetes API inside a fully isolated tenant cluster. Each team gets an isolated tenant cluster with its own K

Every company right now has the same three conversations happening in parallel:

Every company right now has the same three conversations happening in parallel: - Team A wants to call an LLM from their app. They copy an API key into a Secret and ship it.

Learn how NVIDIA Dynamo profiles and deploys AI inference on Kubernetes with a hands-on walkthrough serving Qwen3-0.6B on vCluster Private Nodes.

Deploy NVIDIA Dynamo on Kubernetes using vCluster Private Nodes. This step-by-step guide covers the GPU Operator, DynamoGraphDeploymentRequest, profiling, auto-sizing, and serving Qwen3-0.6B with an OpenAI-compatible API.

AI infrastructure is quickly becoming a platform problem, not just a model problem.

AI infrastructure is quickly becoming a platform problem, not just a model problem. Running one model on one GPU is useful for a demo, but it does not answer the questions platform teams face in practice:

Introducing NVIDIA AI Cluster Runtime and a hands-on deep dive using AICR with vCluster on a private A100 node.

AICR helps generate and validate GPU cluster configurations from the hardware you actually have. This guide introduces NVIDIA AI Cluster Runtime, then walks through resolving recipes, deploying Helm bundles, and validating conformance on a private A100 node with vCluster.

A Ray Serve inference endpoint as an NVCF Helm function, running in a vCluster tenant cluster.

NVIDIA Cloud Functions (NVCF) is the function platform behind NVIDIA's GPU cloud, and it is now open source. For AI Clouds and platform teams building GPU infrastructure, it is one of the more interesting pieces of the stack to study: it tu

Use Kubernetes NetworkPolicy and CiliumNetworkPolicy to block cross-tenant traffic and enforce Tenant Isolation inside a shared Cilium-powered vind Control Plane Cluster.

Block cross-tenant communication in shared vCluster environments using Kubernetes NetworkPolicy and Cilium eBPF-powered security policies step by step.

Deploy multiple Tenant Clusters inside a vind Control Plane Cluster and show how each one transparently inherits host Cilium networking without configuring any CNI of its own.

Learn how Tenant Clusters inside a Cilium-powered vind cluster automatically inherit host networking with zero in-tenant CNI configuration. Deploy workloads, verify IPAM, and test cross-tenant communication.

How vCluster helps sovereign AI Cloud operators strengthen tenant isolation, incident evidence, backup and recovery, and access control for NIS2 readiness.

AI Clouds are directly in scope under NIS2 as cloud computing and data centre service providers. See how vCluster helps operators accelerate readiness with scoped Tenant Clusters, auditable isolation, recovery patterns, and access-control boundaries.

How to stream per-tenant usage events into Lago, Stripe Meters, Metronome, or your own billing adapter. Without building

How to stream per-tenant usage events into Lago, Stripe Meters, Metronome, or your own billing adapter. Without building the pipeline from scratch. The Problem

Accelerating DORA readiness for sovereign AI clouds with scalable tenant isolation, recovery, and audit-ready architectures

DORA introduces strict operational resilience requirements for AI clouds serving financial institutions. This guide explores key challenges and how vCluster helps teams implement tenant isolation, recovery, and audit-ready architectures without infrastructure sprawl.

Accelerating SecNumCloud 3.2 readiness for AI sovereign clouds with scalable tenant isolation and audit-ready architecture

SecNumCloud 3.2 introduces strict requirements for AI sovereign clouds, from tenant isolation to legal sovereignty. This guide explains the key challenges and how vCluster helps platform teams accelerate compliance with scalable, auditable, and efficient Kubernetes architectures.

We just shipped snapshot and restore for vCluster's Docker driver (vind) in v0.34.0. This means you can capture your ent

We just shipped snapshot and restore for vCluster's Docker driver (vind) in v0.34.0. This means you can capture your entire local Kubernetes cluster -- every deployment, every service, every configmap, every PVC with its data -- into a sing

GPU hardware loses value fast. The real competitive advantage in AI infrastructure isn't better chips — it's how quickly you can start monetizing them.

NVIDIA H100 GPUs that sold for $40,000 at launch are already appearing on secondary markets for around $6,000. For organizations building AI infrastructure platforms, every month of delay means compounding losses from depreciation, engineering burn, and missed revenue. The question isn't which GPU to buy — it's how fast you can get your platform to production.

Metrics, dashboards, alerting, and long-term storage, deployed on a multi-node Kubernetes cluster that runs entirely in Docker

Observability isn't optional, it's foundational. This guide walks through deploying Prometheus, Grafana, Thanos, and RustFS on a local multi-node Kubernetes cluster powered by vCluster. No cloud account required. No VMs. Just Docker and a few commands.

Most GPU data centers take months to operationalize. vMetal gets you from rack to running cluster in minutes.

Buying GPUs is the easy part. Operating them like a cloud platform is what separates Neoclouds that launch from ones that stall. vMetal automates the full bare metal lifecycle — discovery, provisioning, cluster attachment — so you can start delivering GPU capacity immediately.

A web dashboard for all your local vind clusters with projects, team management, role-based access, and automation through access keys

The CLI is great for daily dev work. But when you need to see all your clusters at a glance, organize them into projects, manage team access, or hand off a demo environment, the vCluster Platform UI takes vind from a better KinD to a real cluster management platform. Day 7 of the 7 Days of vind series.

Pause clusters and resume where you left off, pull locally built images in 45ms, install Cilium, and map custom ports to your local cluster

With KinD, you delete your cluster at the end of the day and recreate it tomorrow. With vind, you pause it and resume exactly where you left off. Deployments, services, PVCs, all still there. Day 6 covers sleep/wake, registry proxy, custom CNI, and more.

Drop setup-kind from your GitHub Actions and get built-in registry proxy, automatic log export, and multi-cluster CI workflows with setup-vind

If you're using setup-kind in GitHub Actions, you're still loading images manually and missing automatic log exports. setup-vind is a drop-in replacement with built-in registry proxy, automatic artifact export, and multi-cluster support. Day 5 of the 7 Days of vind series.

Run your control plane locally in Docker, join a real cloud VM as a worker node over VPN, and schedule pods across both

Local Kubernetes tools stop at your laptop. vind doesn't. Join a GCP Compute Engine instance as a real worker node to your local cluster over an encrypted VPN tunnel. Test GPU workloads, mixed architectures, and hybrid setups. Day 4 of the 7 Days of vind series.

Create a multi-node local cluster in Docker and test pod distribution, node drains, affinity, and anti-affinity, just like production

Single-node clusters can't test scheduling. With vind, spin up a 4-node cluster in Docker, deploy across workers, drain nodes, and test affinity rules, real Kubernetes behavior on your laptop. Day 3 of the 7 Days of vind series.