{"resourceId":"nvidia-kubernetes-engine-recipes-public-preview-20261110","versions":[{"version":"external-34a70bbc17fc01b4903f1ff3efc4036a05eca7dcaac81313ded13402745afe28","resource":{"id":"nvidia-kubernetes-engine-recipes-public-preview-20261110","title":"Manage Kubernetes Clusters for GPU Infrastructure with NVIDIA Kubernetes Engine","organization":"NVIDIA","sector":"AI factory / GPU infrastructure software","geography":"Global","publishedAt":"November 10, 2026","publicationDate":"2026-11-10","eventDate":"2026-11-10","sourceName":"NVIDIA Technical Blog","sourceLabel":"NVIDIA Technical Blog (vendor documentation)","sourceUrl":"https://developer.nvidia.com/blog/manage-kubernetes-clusters-for-gpu-infrastructure-with-nvidia-kubernetes-engine/","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["infrastructure","operating-model"],"finding":"NVIDIA announced two open source projects, NVIDIA Kubernetes Engine (NVKE) and NVIDIA Kubernetes Recipes (NVKR), for running managed Kubernetes services on GPU infrastructure. NVKR builds version-pinned node and container images and records build inputs; NVKE provisions and manages clusters through a REST API and controllers, with servers supplied by NVIDIA Infra Controller. Both are Apache 2.0, in public preview, and NVIDIA states they are not yet recommended for production workloads. NVKE is described as a Certified Kubernetes AI Platform and Installer, conformance tested on Kubernetes 1.36. Firmus and SK Telecom are integrating them; Mirantis, Rafay, Spectro Cloud, OpenNebula Systems, vCluster and Netris are incorporating components.","sledRelevance":"Limited and indirect. This targets AI factory operators and cloud providers, not typical agencies or campuses. It matters to SLED teams only where a state, university or research-computing center operates its own GPU cluster or buys capacity from a provider that adopts it.","evidence":"Vendor technical post with a walkthrough; no independent test, customer deployment or performance data. Partner integrations are stated as in progress.","architectureImplications":"Management cluster seeded by nvkectl, workload clusters with separate control planes and node pools, bare-metal lifecycle via NICo and Cluster API, attested SPIFFE node identity, and an OpenTelemetry telemetry chart. Teams can adopt NVKE and NVKR together or separately.","governanceImplications":"Version-pinned, recorded build inputs support reproducibility and change control. Preview status means APIs may change and production support is not offered.","securityPrivacyImplications":"Signed, version-pinned images and attested node identity are positive design points as described; the post gives no security assessment, and tenant isolation behavior is not tested in the source.","caveats":"Vendor-authored; public preview; not recommended for production; walkthrough prerequisites are substantial (multiple CPU servers, GPU pool, NICo installation, build host).","streamIds":["nvidia"],"roles":{"sales":"This is an operator-side story, so qualify carefully. It applies to institutions or providers that run GPU capacity as a service, such as research computing centers or regional consortia, not to ordinary application buyers. Use it to ask how they plan to offer Kubernetes access to GPUs and who supports it, while stating plainly that it is in preview and not recommended for production. Do not position it as a ready managed service. This is interpretation based on the vendor's own preview statement.","engineering":"Engineering should read the walkthrough prerequisites first: three CPU servers for management, three for a workload control plane, a GPU pool, a NICo installation, a build host and a registry. Evaluate NVKE on the single-machine quickstart before committing hardware. Check conformance claims, upgrade and recovery behavior, tenant isolation and telemetry export against your standards. Because APIs may change in preview, isolate dependencies and plan for rework before any production use.","delivery":"Delivery should scope a lab proof of concept, not a production rollout. Work includes building a node image, standing up the management plane, and provisioning one test cluster. Dependencies are NICo capacity, networking, registry and image hosting. Assign an infrastructure owner and an SRE-style on-call plan before expansion. Acceptance should rest on repeatable builds, successful upgrade and recovery tests, and documented known limitations from the preview status page."},"roleDetails":{"sales":{"customerProblem":"Institutions running or buying GPU capacity need a supportable way to offer Kubernetes access to tenants.","stakeholders":["Research computing director","Infrastructure or platform lead","Cloud or colocation provider","Procurement"],"discoveryQuestions":["Do you operate GPU clusters or buy them as a service?","What Kubernetes management tools do you use today?","What support level do you require for production?"],"valueHypothesis":"Reproducible images and an API-driven cluster lifecycle may reduce manual GPU cluster operations once the projects mature.","potentialEngagement":"Lab-scale evaluation and architecture review of NVKE and NVKR against current cluster management.","unsupportedClaims":["That NVKE is production ready","That partner integrations are shipping","Any cost or staffing savings"]},"engineering":{"fit":"Fits operators building managed Kubernetes on bare-metal GPU servers, especially with NICo.","architectureIntegration":"nvkectl seed cluster forms a management cluster; NVKE REST API and controllers create workload clusters; NICo supplies servers via CAPNICo; NVKR supplies images.","prerequisites":["Three CPU servers for management","Three CPU servers plus GPU pool for a workload cluster","NICo with enrolled capacity","Build host with Docker and KVM","Registry and HTTP image server"],"constraints":["Public preview; APIs may change","Not recommended for production","Heavy hardware prerequisites"],"security":"Verify image signing, SPIFFE attestation and tenant isolation yourself; no assessment is provided in the post.","proofOfValue":["Run the single-machine quickstart","Build a node image and reproduce it","Provision a test cluster and exercise upgrade and recovery"]},"delivery":{"work":"Lab proof of concept: image build, management plane, one test workload cluster.","dependencies":["NICo capacity","Network and registry","Image hosting"],"ownership":"Infrastructure team lead with named on-call owner.","skillsAdoption":"Needs Kubernetes, Cluster API, bare-metal and image-build skills.","governanceCheckpoints":["Review preview limitations before hardware commitment","Change control for version pins","Upgrade and recovery test review"],"acceptanceCriteria":["Proposed: reproducible node image rebuild","Proposed: successful cluster provision, upgrade and recovery test"],"risks":["API changes during preview","Hardware prerequisites","Skills gap"]}},"retrievedAt":"2026-11-10T21:59:07-06:00","enrichedAt":"2026-11-10T22:02:23-06:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Bare-metal GPU Kubernetes operations demand scarce platform engineering skills; open source documentation and a local quickstart lower the learning barrier but not the staffing need.","procurementImplications":"Apache 2.0 licensing and open code allow inspection, but preview status means no production support commitment; contracts with any provider adopting NVKE should state support, versioning and exit terms.","operatingModelImplications":"Suggests a platform-team model where an AI factory operator runs the management plane and tenants consume clusters through APIs, with release sets controlling validated Kubernetes versions.","sourceVerification":{"openedUrl":"https://developer.nvidia.com/blog/manage-kubernetes-clusters-for-gpu-infrastructure-with-nvidia-kubernetes-engine/","referenceExcerpt":"NVKE and NVKR are in public preview and are not yet recommended for production workloads.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}