{"slug": "nvidia-kubernetes-engine-nvke", "title": "Nvidia Kubernetes Engine (NVKE)", "summary": "NVIDIA released NVIDIA Kubernetes Engine (NVKE) as a preview, a management plane that platform teams install to run Kubernetes workload clusters on their own NVIDIA GPU infrastructure, with each cluster getting its own control plane, etcd and node pools. NVKE provisions machines through NVIDIA Infra Controller (NICo), the sole compatible infrastructure provider at preview, and nodes attest through SPIFFE/SPIRE using their TPM before receiving join credentials. NVIDIA warns the preview is not recommended for production workloads: API authentication is off and the API uses plain HTTP after installation, the management cluster cannot upgrade, scale or replace itself, there is no management-plane upgrade or backup/restore procedure, and FDE node images provide no data-at-rest confidentiality.", "body_md": "**Run a managed Kubernetes service on your own NVIDIA GPU infrastructure.**\nNVIDIA Kubernetes® Engine (NVKE) provides the management plane that platform teams install to create and\noperate Kubernetes clusters for their customers. Each workload cluster has its\nown Kubernetes control plane, etcd and node pools. You operate the NVKE management\nplane; your customers use their clusters through the Kubernetes API and `kubectl`.\n\n**Preview:** NVKE is available as a preview. APIs, configuration and behavior\ncan change between releases, and we do not recommend NVKE for production\nworkloads. See [Preview status and known limitations](https://github.com/dsx-ai-factory/nvke/blob/main/docs/overview/preview-status.md).\n\n[Get started](#getting-started) · [Documentation](https://github.com/dsx-ai-factory/nvke/blob/main/docs/README.md) ·\n[Contribute](https://github.com/dsx-ai-factory/nvke/blob/main/CONTRIBUTING.md) · [Get help](https://github.com/dsx-ai-factory/nvke/blob/main/SUPPORT.md)\n\n| Your goal | Start here | What you get | \n|---|---|---|\n| **Evaluate without GPU hardware** | [Local quickstart](https://github.com/dsx-ai-factory/nvke/blob/main/docs/get-started/quickstart-local.md) | A development management plane and a workload cluster on simulated machines. No NVIDIA account is required. | \n| **Install on your infrastructure** | [Plan your installation](https://github.com/dsx-ai-factory/nvke/blob/main/docs/operate/plan-installation.md) | The prerequisites and sequence for preparing nodes, bootstrapping a management cluster and installing its services. | \n| **Build or contribute** | [Build from source](#build-from-source) | The NVKE binaries and the contributor development loop. | \n\nThe local quickstart runs on a **Linux host or VM, amd64 or arm64**, with at least\n**6 CPUs, 12 GiB of free memory and 100 GiB of free disk**. It requires Docker,\nvirtual TPM support and the tools listed in the guide. Follow its setup through\nworkload-cluster creation and confirm the cluster's nodes report `Ready`.\nOn macOS, use the [local site guide](https://github.com/dsx-ai-factory/nvke/blob/main/docs/get-started/local-site.md), which runs\nthe evaluation inside a Colima virtual machine.\n\nFor a real installation, you supply the machines, networking, node images and\nregistries, and configure access to **NVIDIA Infra Controller (NICo)**. NICo is\nthe sole compatible infrastructure provider at preview. The\n[platform and version reference](https://github.com/dsx-ai-factory/nvke/blob/main/docs/overview/supported-platforms.md) lists\nvalidated combinations and pinned dependencies.\n\n- **Creates independent workload clusters.** Request clusters through the NVKE\nAPI or`nvkectl` ; NVKE provisions machines through NICo and forms Kubernetes\nclusters with their own control planes.\n- **Manages workload cluster lifecycle.** Create, scale and delete node pools,\nupgrade Kubernetes and replace nodes. See the[node pool guide](https://github.com/dsx-ai-factory/nvke/blob/main/docs/workload-clusters/node-pools.md) for the manual steps and preview limits on replacement and upgrades.\n- **Establishes node identity.** Nodes attest through SPIFFE/SPIRE using their\nTPM before receiving join credentials. Encryption of etcd data is available\nper cluster.\n- **Connects builds to deployment.** Its companion project,[NVIDIA Kubernetes Recipes (NVKR)](https://github.com/dsx-ai-factory/nvkr) ,\nbuilds node images and NVKE components from pinned inputs. NVKE deploys those\nartifacts; NVKR owns their build and release evidence.\n\nThe preview needs explicit operator configuration: API authentication is off\nand the API uses plain HTTP after installation. Configure\n[API authentication and protected access](https://github.com/dsx-ai-factory/nvke/blob/main/docs/operate/api-authentication.md)\nbefore admitting callers. The management cluster does not upgrade, scale or\nreplace itself, and there is no management-plane upgrade or backup/restore\nprocedure. FDE node images are for evaluation and provide no data-at-rest\nconfidentiality. Read the [known limitations](https://github.com/dsx-ai-factory/nvke/blob/main/docs/overview/preview-status.md)\nbefore choosing an evaluation or installation path.\n\nThe diagram shows the NVKE product family, how it sits on NICo and your GPU machines, and which components talk to each other.\n\nNVKE is not a Kubernetes distribution and not a Cluster API flavour. It is a\n**self-hosting factory for Kubernetes clusters**: a throwaway seed cluster\nstands up a management cluster, and the management plane on\nit then provisions and lifecycle-manages many isolated workload clusters\non demand.\n\nAn installation has two kinds of Kubernetes cluster:\n\n- The **management cluster** runs the NVKE management plane and infrastructure\ncontrollers.`nvkectl` forms it using a temporary seed cluster; installing the\nmanagement-plane services is a separate step.\n- **Workload clusters** run customer applications. The management plane creates\nand operates them through NICo, Cluster API and the node agent.\n\nRead [What is NVKE](https://github.com/dsx-ai-factory/nvke/blob/main/docs/overview/what-is-nvke.md) for the operator workflow and\n[Architecture](https://github.com/dsx-ai-factory/nvke/blob/main/docs/architecture.md) for the implementation and data flows.\n\nUse **Go 1.27.3**, the version pinned in [go.work](https://github.com/dsx-ai-factory/nvke/blob/main/go.work), and **GNU Make 4 or later**. Public Go modules must be\nreachable on the first build. [Flox](https://flox.dev/) is optional and supplies\nthe repository's development tools; on macOS, it also supplies GNU Make 4.\nSee [Development](https://github.com/dsx-ai-factory/nvke/blob/main/docs/development.md) for toolchain setup.\n\n```\ngit clone https://github.com/dsx-ai-factory/nvke.git\ncd nvke\nmake build\n```\n\nThis builds the workspace and writes nine binaries to `bin/`: `agent`,\n`updater`, `controller`, `nvkectl`, `capd-reboot-controller`,\n`nvke-os-image-check`, `management-engine`, `tenant-api` and `operator-api`. Building does not start a\nmanagement plane or create a cluster. The node agent runs on Linux.\n\nThe [workspace guide](https://github.com/dsx-ai-factory/nvke/blob/main/docs/go-workspace.md) explains the modules and outputs.\nFor local services and test environments, see [Local development setup](https://github.com/dsx-ai-factory/nvke/blob/main/docs/getting-started.md).\nBefore submitting a change, follow [CONTRIBUTING.md](https://github.com/dsx-ai-factory/nvke/blob/main/CONTRIBUTING.md) and run\n`make preflight`; it selects checks for the modules you changed.\n\n| Task | Guide | \n|---|---|\n| Prepare nodes and bootstrap the management cluster | [Bootstrap the management cluster](https://github.com/dsx-ai-factory/nvke/blob/main/docs/operate/bootstrap-management-cluster.md) | \n| Install NVKE services | [Install the management plane from the deploy path](https://github.com/dsx-ai-factory/nvke/blob/main/deploy/README.md) | \n| Build node images and supply release artifacts | [Get NVKE artifacts from NVKR](https://github.com/dsx-ai-factory/nvke/blob/main/docs/operate/build-node-images.md) | \n| Create and manage customer clusters | [Create a workload cluster](https://github.com/dsx-ai-factory/nvke/blob/main/docs/workload-clusters/create.md) ,[Manage node pools](https://github.com/dsx-ai-factory/nvke/blob/main/docs/workload-clusters/node-pools.md) | \n| Give a customer access | [Access a workload cluster](https://github.com/dsx-ai-factory/nvke/blob/main/docs/workload-clusters/access.md) | \n| Inspect cluster health and telemetry | [Observe NVKE and its clusters](https://github.com/dsx-ai-factory/nvke/blob/main/docs/operate/observe.md) | \n\nThe [documentation index](https://github.com/dsx-ai-factory/nvke/blob/main/docs/README.md) links the remaining guides and references.\n\nContributions, bug reports and documentation improvements are welcome.\nStart with [CONTRIBUTING.md](https://github.com/dsx-ai-factory/nvke/blob/main/CONTRIBUTING.md); open an issue before substantial\nwork so maintainers can agree on the approach.\n\n- **Questions:**[GitHub Discussions](https://github.com/dsx-ai-factory/nvke/discussions/categories/q-a) .\n- **Chat:** the`#nvke` channel in the DSX AI Factory Slack.[Get an invite](https://inviter.co/dsx-ai-factory) .\n- **Bugs and feature requests:**[GitHub Issues](https://github.com/dsx-ai-factory/nvke/issues) .\n- **Security vulnerabilities:** follow[SECURITY.md](https://github.com/dsx-ai-factory/nvke/blob/main/SECURITY.md) and report privately.\n- **Community:**[Code of Conduct](https://github.com/dsx-ai-factory/nvke/blob/main/CODE_OF_CONDUCT.md) ,[Governance](https://github.com/dsx-ai-factory/nvke/blob/main/GOVERNANCE.md) ,[Maintainers](https://github.com/dsx-ai-factory/nvke/blob/main/MAINTAINERS.md) and[Support](https://github.com/dsx-ai-factory/nvke/blob/main/SUPPORT.md) .\n- **Releases:**[release process](https://github.com/dsx-ai-factory/nvke/blob/main/RELEASE.md) .\n- **CI health:** the[CI dashboard](https://dsx-ai-factory.github.io/nvke/) shows the last thirty runs of every workflow and refreshes hourly.\n\nKubernetes® is a registered trademark of The Linux Foundation in the United States and other countries. NVIDIA and the NVIDIA logo are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with which they are associated.\n\nApache License 2.0. See [LICENSE](https://github.com/dsx-ai-factory/nvke/blob/main/LICENSE) and [NOTICE](https://github.com/dsx-ai-factory/nvke/blob/main/NOTICE).\n\nCopyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.", "url": "https://wpnews.pro/news/nvidia-kubernetes-engine-nvke", "canonical_source": "https://github.com/dsx-ai-factory/nvke", "published_at": "2026-11-06 03:16:26+00:00", "updated_at": "2026-11-06 03:20:01.584237+00:00", "lang": "en", "topics": ["ai-infrastructure", "mlops", "developer-tools", "ai-tools"], "entities": ["NVIDIA", "NVIDIA Kubernetes Engine", "NVIDIA Infra Controller", "NVIDIA Kubernetes Recipes", "SPIFFE", "SPIRE", "nvkectl"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/nvidia-kubernetes-engine-nvke", "markdown": "https://wpnews.pro/news/nvidia-kubernetes-engine-nvke.md", "text": "https://wpnews.pro/news/nvidia-kubernetes-engine-nvke.txt", "jsonld": "https://wpnews.pro/news/nvidia-kubernetes-engine-nvke.jsonld"}}