11-19 更新 实时
首页/ 算力/ 正文

NVIDIA 开源 NVKE、NVKR:GPU 上的托管 K8s

算力 来源:MarkTechPost 2026-11-17 18:37 0 阅读 机器翻译

NVIDIA 发布了 2 个用于在 GPU 基础设施上运行托管 Kubernetes 的开源项目:NVIDIA Kubernetes Engine(NVKE)和 NVIDIA Kubernetes Recipes(NVKR)。二者近期作为 NVIDIA DSX AI Factory Platform 的组件对外公布。NVKE 是托管 Kubernetes 服务的控制平面。NVKR 负责构建这些集群所依赖的操作系统镜像、容器镜像和软件包。两者均以 Apache 2.0 许可发布,目前处于公开预览阶段。

NVKE 是一款开源软件,AI 工厂安装它之后,便可在自有 NVIDIA GPU 基础设施上运行托管 Kubernetes 服务。该仓库称其为“Kubernetes 集群的自托管工厂”,而不是一个发行版。它在管理集群上暴露 REST API 和控制器。客户获得标准集群并使用 kubectl。运维人员通过 NVKE API 或 nvkectl CLI 对其进行管理。

NVKE 使用上游 Kubernetes、kubeadm 和 Cluster API。它用 Go 编写,运行时并不依赖 NVKR。

NVKR 是一套带有声明式 Definition 的构建系统。它可生成面向 amd64 和 arm64 的可执行文件、容器镜像、Helm chart 和节点镜像。“Recipe”定义了经过验证的软件版本与 GPU 代际组合。托管工作流会为固定摘要(digest)的 OCI 引用附加签名、SBOM 和来源信息。NVKR 也可为其他 Kubernetes 管理系统构建产物。

NVIDIA 技术博客的实操指南描述了 5 个阶段:

节点必须先出示经过认证的 SPIFFE 身份,才能获得加入凭证。README 称 SPIRE 为此使用了 TPM。可按集群启用 etcd 加密。一个基于 OpenTelemetry 的 Helm chart 通过 OTLP 导出日志、指标和追踪数据。

在工作负载集群内部,有 3 个配套项目可接入:用于 GPU 运行时的 AI Cluster Runtime(AICR)、用于节点软件包的 NodeWright,以及用于故障修复自动化的 NVSentinel。

NVIDIA 云合作伙伴 Firmus 和 SK Telecom 正在将这两个项目集成到 AI 工厂中。Mirantis、Rafay、Spectro Cloud、OpenNebula Systems 和 vCluster 正在将相关组件加入其托管 Kubernetes 产品。Netris 正在为其网络自动化启用 NVKE。

本地快速入门可运行在 1 台 Linux 主机或虚拟机上,支持 amd64 或 arm64。它需要至少 6 个 CPU、12 GiB 可用内存、100 GiB 可用磁盘空间,以及 Docker 和虚拟 TPM 支持。无需 GPU,也无需 NVIDIA 账号。正式安装则需要一个已完成容量注册的 NICo 部署、一台带 KVM 的构建主机、一个 HTTP 服务器和一个镜像仓库(registry)。

NVKR 目前尚无通过认证的 Release。

请查看 NVKE 仓库、NVKR 仓库、文档和技术博客。你可以通过我们的 GPULeaderboard.com Chrome 扩展在这里查看部署选项和规格。

需要与我们合作推广你的 GitHub 仓库、Hugging Face 页面、产品发布或网络研讨会等?欢迎与我们联系。

《NVIDIA Open-Sources NVKE and NVKR for Managed Kubernetes on GPU Infrastructure》一文最初发布于 MarkTechPost。

译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。

查看英文原文(NVIDIA Open-Sources NVKE and NVKR for Managed Kubernetes on GPU Infrastructure)

NVIDIA has released 2 open-source projects for running managed Kubernetes on GPU infrastructure: NVIDIA Kubernetes Engine (NVKE) and NVIDIA Kubernetes Recipes (NVKR). Both were recently announced as components of the NVIDIA DSX AI Factory Platform. NVKE is the control plane for a managed Kubernetes service. NVKR builds the OS images, container images and packages those clusters depend on. Both ship under Apache 2.0 and are in public preview.

TL;DR

  • Size / footprint: 3 CPU servers for the management cluster, 3 CPU servers per workload control plane, plus a pool of GPU worker nodes. A local evaluation needs 6 CPUs, 12 GiB RAM, 100 GiB disk and no GPU.
  • Runs on: your own NVIDIA GPU bare-metal servers, provisioned through NVIDIA Infra Controller (NICo). Node images build for amd64 and arm64.
  • Performance: no throughput or provisioning-time benchmarks are published. Quality signal is certification, not speed.
  • Best: Certified Kubernetes AI Platform and Certified Kubernetes Installer, conformance tested on Kubernetes 1.36.
  • Worst: after install, the NVKE API serves plain HTTP on port 30080 with authentication off, and there is no management-plane upgrade or backup procedure.
  • Bottom line: best thing: open, attested, upstream Kubernetes for GPU clouds. Worst thing: preview-grade security and lifecycle gaps; not for production yet.

What is NVIDIA Kubernetes Engine (NVKE)?

NVKE is open-source software that an AI factory installs to run a managed Kubernetes service on its own NVIDIA GPU infrastructure. The repo calls it ‘a self-hosting factory for Kubernetes clusters,’ not a distribution. It exposes a REST API and controllers on a management cluster. Customers get standard clusters and use kubectl. The operator manages them through the NVKE API or the nvkectl CLI.

NVKE uses upstream Kubernetes, kubeadm and Cluster API. It is written in Go and has no runtime dependency on NVKR.

What does NVIDIA Kubernetes Recipes (NVKR) do?

NVKR is a build system with declarative Definitions. It produces executables, container images, Helm charts and node images for amd64 and arm64. ‘Recipes’ define validated combinations of software versions and GPU generations. Hosted workflows attach signatures, SBOMs and provenance to digest-pinned OCI references. NVKR can also build artifacts for other Kubernetes management systems.

How does the NVKE architecture work?

The NVIDIA technical blog walkthrough describes 5 stages:

  1. Bootstrap: nvkectl bootstrap site starts a temporary single-node kind cluster on a bootstrap host.
  2. Management cluster: NVKE requests 3 servers from NICo, which boots them from the node image. The temporary cluster is then deleted.
  3. Machine provisioning: NICo handles bare-metal lifecycle. CAPNICo, a Cluster API provider, connects NICo to NVKE.
  4. Release sets: operators publish a machine image, a cluster release and a node pool release. Customers pick a release by name.
  5. Workload clusters: each customer cluster has its own control plane, etcd and node pools. NVKE installs OVN-Kubernetes and NodeLocal DNSCache.

A node must present an attested SPIFFE identity before it gets join credentials. The README says SPIRE uses TPM for this. Per-cluster etcd encryption is available. An OpenTelemetry-based Helm chart exports logs, metrics and traces over OTLP.

Inside workload clusters, 3 companion projects plug in: AI Cluster Runtime (AICR) for the GPU runtime, NodeWright for node packages and NVSentinel for break-fix automation.

How does NVKE compare with Gardener and Kamaji?

NVKE + NVKR (NVIDIA)Gardener (SAP)Kamaji (Clastix)
What it isManaged Kubernetes management plane plus artifact builderKubernetes clusters as a serviceHosted control plane manager
Control plane modelDedicated per cluster: own control plane, etcd, node poolsRuns as pods in seed clustersRuns as pods; shared Datastore possible
Target infrastructureNVIDIA GPU bare metal via NICoAWS, Azure, GCP, OpenStack, Alicloud (conformance table); extensibleWorks alongside Cluster API providers
LicenseApache 2.0Apache 2.0Apache 2.0
ConformanceCertified Kubernetes AI Platform and Installer; tested on 1.36Certified up to v1.35Described as CNCF-compliant
Node identitySPIFFE/SPIRE attestation with TPMNot disclosedNot disclosed
Signed build pipelineYes (NVKR: signatures, SBOMs, provenance)Not disclosedNot disclosed
MaturityPublic preview, not for productionSelf-described production-readyNot disclosed

Who is adopting NVKE and NVKR?

NVIDIA Cloud Partners Firmus and SK Telecom are integrating both projects into AI factories. Mirantis, Rafay, Spectro Cloud, OpenNebula Systems and vCluster are adding components to managed Kubernetes offerings. Netris is enabling NVKE for its network automation.

How do you try NVKE?

A local quickstart runs on 1 Linux host or VM, amd64 or arm64. It needs at least 6 CPUs, 12 GiB free memory, 100 GiB free disk, Docker and virtual TPM support. No GPU or NVIDIA account is required. A real install needs a NICo deployment with enrolled capacity, a build host with KVM, an HTTP server and a registry.

NVKR has no qualified Release yet.

Key Takeaways

  • NVKE manages clusters; NVKR builds and signs what they run.
  • Both are Apache 2.0, in public preview, not for production.
  • Conformance tested on Kubernetes 1.36, with TPM-attested node identity.
  • Default API is HTTP on port 30080 with auth off.
  • Local evaluation needs 6 CPUs, 12 GiB RAM and no GPU.


Check out the NVKE repo, NVKR repo, docs and technical blog. You can check the deployment options and specs here through our GPULeaderboard.com chrome extension.

Also, feel free to follow us on Twitter and don’t forget to join our 150k+ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well.

Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us

The post NVIDIA Open-Sources NVKE and NVKR for Managed Kubernetes on GPU Infrastructure appeared first on MarkTechPost.

正文保留英文原文,标题与摘要由机器翻译。

本文由采集器自动同步

原文发布于 MarkTechPost(NVIDIA Open-Sources NVKE and NVKR for Managed Kubernetes on GPU Infrastructure),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。

阅读原文 ↗