Skip to content

docs(conformance): update CNCF evidence for multi-platform and training#425

Merged
yuanchen8911 merged 1 commit intoNVIDIA:mainfrom
yuanchen8911:docs/cncf-evidence-multi-platform
Mar 18, 2026
Merged

docs(conformance): update CNCF evidence for multi-platform and training#425
yuanchen8911 merged 1 commit intoNVIDIA:mainfrom
yuanchen8911:docs/cncf-evidence-multi-platform

Conversation

@yuanchen8911
Copy link
Copy Markdown
Contributor

Summary

  • Add 3 SHOULD requirements (driver_runtime_management, gpu_sharing, virtualized_accelerator) to PRODUCT.yaml
  • Add Kubeflow Trainer (training) evidence to robust-operator.md alongside Dynamo (inference)
  • Add GKE cluster autoscaler evidence to cluster-autoscaling.md alongside EKS ASG
  • Redact cloud-provider artifacts (node names, DNS, project IDs) from all evidence files except cluster-autoscaling.md for platform-neutral presentation
  • Update all evidence doc headers to platform-neutral format
  • Fix PRODUCT.yaml: add k8sConformanceUrl, update productLogoUrl to SVG, set platformName to "NVIDIA AICR-Enabled Kubernetes Platform"
  • Remove Karpenter references from docs and code
  • Update all Kubernetes version references to v1.35
  • Update submission/README.md and index files for consistency

CNCF submission readiness

  • All 9 MUST requirement IDs: Implemented with evidence
  • All 3 SHOULD requirement IDs: Implemented with evidence/docs
  • k8sConformanceUrl set to EKS k8s-conformance entry (AICR validated on multiple conformant platforms)
  • productLogoUrl updated to raw SVG from CNCF landscape
  • Platform-specific details confined to cluster-autoscaling.md (by design)

Test plan

  • make test passes (Go code change in pkg/evidence/requirements.go)
  • Verify markdown renders correctly on GitHub
  • Verify PRODUCT.yaml passes CNCF validator schema check

@yuanchen8911 yuanchen8911 requested a review from a team as a code owner March 18, 2026 16:53
@yuanchen8911 yuanchen8911 requested review from dims March 18, 2026 16:55
dims
dims previously approved these changes Mar 18, 2026
@yuanchen8911 yuanchen8911 force-pushed the docs/cncf-evidence-multi-platform branch 2 times, most recently from aa732f4 to 1de9594 Compare March 18, 2026 17:24
Update CNCF AI Conformance evidence and submission for platform-neutral
presentation with multi-intent coverage.

- Set platformName to "NVIDIA AI Cluster Runtime" matching README
- Add 3 SHOULD requirements (driver_runtime_management, gpu_sharing,
  virtualized_accelerator) to PRODUCT.yaml
- Add Kubeflow Trainer (training) evidence to robust-operator.md alongside
  Dynamo (inference)
- Add GKE cluster autoscaler evidence to cluster-autoscaling.md alongside
  EKS ASG
- Redact cloud-provider artifacts (node names, DNS, project IDs) from all
  evidence files except cluster-autoscaling.md
- Update headers to platform-neutral format across all evidence docs
- Update submission/README.md, evidence/index.md, and cncf/index.md for
  consistency
- Fix PRODUCT.yaml: k8sConformanceUrl, productLogoUrl (SVG)
- Remove Karpenter references from docs and code
- Update Kubernetes version references to v1.35

Signed-off-by: Yuan Chen <yuanchen97@gmail.com>
@yuanchen8911 yuanchen8911 force-pushed the docs/cncf-evidence-multi-platform branch from 1de9594 to 06a92e6 Compare March 18, 2026 17:41
@yuanchen8911 yuanchen8911 merged commit ff9c66d into NVIDIA:main Mar 18, 2026
25 checks passed
xdu31 pushed a commit to xdu31/aicr that referenced this pull request Mar 24, 2026
…ng (NVIDIA#425)

Signed-off-by: Yuan Chen <yuanchen97@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants