Oriel — AI & technology worth noticing, every day.

Manage Kubernetes Clusters for GPU Infrastructure with NVIDIA Kubernetes Engine

Web page · https://developer.nvidia.com/blog/manage-kubernetes-clusters-for-gpu-infrastructure-with-nvidia-kubernetes-engine

NVIDIA is announcing two open source projects for building and operating managed Kubernetes services on GPU infrastructure—NVIDIA Kubernetes Engine (NVKE) and...

First seen by Oriel on .

AI summary

NVIDIA introduces open source NVKE and NVKR in public preview for building reproducible software artifacts and operating managed Kubernetes services on GPU infrastructure.

  • NVKR creates version-pinned node images, binaries, container images, and Helm charts, recording build inputs so teams can identify and reproduce deployed software.
  • nvkectl forms a management cluster through a temporary kind seed cluster, while NICo provisions bare-metal machines through the CAPNICo Cluster API provider.
  • Each customer workload cluster receives a separate Kubernetes control plane, etcd, and node pools, with published release sets abstracting images and cloud-init configuration.
  • Nodes must present attested SPIFFE identities before receiving join credentials; OpenTelemetry-based telemetry exports logs, metrics, and traces through OTLP.
  • NVKE is conformance tested on Kubernetes 1.36, but preview APIs, configuration, and behavior may change, and production workloads are not yet recommended.

#kubernetes #gpu infrastructure #nvke #reproducible builds

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
NVIDIA Developer Bloghttps://developer.nvidia.com/blog/?p=124939https://developer.nvidia.com/blog/manage-kubernetes-clusters-for-gpu-infrastructure-with-nvidia-kubernetes-engine/ Read discussion