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
| Source | External ID | URL |
|---|---|---|
| NVIDIA Developer Blog | https://developer.nvidia.com/blog/?p=124939 | https://developer.nvidia.com/blog/manage-kubernetes-clusters-for-gpu-infrastructure-with-nvidia-kubernetes-engine/ Read discussion |