Skip to content

docs(topology): document NVCF-native MNNVL scheduling primitives - #744

Merged
sbaum1994 merged 1 commit into
mainfrom
estroczynski/docs/legacy-nvlink-sched
Aug 10, 2026
Merged

docs(topology): document NVCF-native MNNVL scheduling primitives#744
sbaum1994 merged 1 commit into
mainfrom
estroczynski/docs/legacy-nvlink-sched

Conversation

@estroz

@estroz estroz commented Aug 10, 2026

Copy link
Copy Markdown
Contributor

TL;DR

This MR documents NVCF's native MNNVL scheduling primitives, in lieu of KAI/Grove usage

Additional Details (optional for docs, build, test, refactor, ci, chore, style, and revert PRs)

For the Reviewer

For QA (optional for docs, build, test, refactor, ci, chore, style, and revert PRs)

Issues

Relates to #598

Checklist

  • I am familiar with the Contributing Guidelines.
  • I have signed off my commits for Developer Certificate of Origin (DCO) compliance.
  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

Summary by CodeRabbit

  • Documentation
    • Documented legacy NVLink partition annotations and their supported workload behavior.
    • Clarified best-effort placement, logical grouping, affinity, and Pod mutation semantics.
    • Added guidance on using gang and topology-aware scheduling for coordinated workload placement.
    • Updated Helm examples with NVLink placement options and scheduler limitations.

Signed-off-by: Eric Stroczynski <estroczynski@nvidia.com>
@estroz
estroz requested review from a team as code owners August 10, 2026 18:31
@estroz
estroz requested a review from vrv3814 August 10, 2026 18:31
@coderabbitai

coderabbitai Bot commented Aug 10, 2026

Copy link
Copy Markdown

Review Change Stack

📝 Walkthrough

Walkthrough

This PR documents legacy NVLink partition annotations for Helm functions. It describes their Pod-affinity behavior, logical grouping, Pod mutations, and best-effort placement limits. It also updates topology scheduling guidance and a multi-node StatefulSet example.

Changes

Legacy NVLink placement documentation

Layer / File(s) Summary
Annotation semantics
docs/user/helm-functions.md
Documents the required-nvlink-domain-index annotation, logical partition behavior, supported placement semantics, required locations, and NVCA mutations to admitted Pods.
Usage guidance and example
docs/user/cluster-management/topology-aware-scheduling.md, examples/function-samples/helmchart-samples/multi-node-helm-function-test/multi-node-test/templates/statefulset.yaml
Documents best-effort placement without KAI Scheduler or Grove and adds corresponding comments to the multi-node StatefulSet example.

Estimated code review effort: 1 (Trivial) | ~5 minutes

Possibly related PRs

  • NVIDIA/nvcf#603: Documents related legacy NVLink annotations and KAI/Grove topology-aware alternatives.

Suggested reviewers: vrv3814, kristinapathak

🚥 Pre-merge checks | ✅ 5
✅ Passed checks (5 passed)
Check name Status Explanation
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check ✅ Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check ✅ Passed Check skipped because no linked issues were found for this pull request.
Description Check ✅ Passed Check skipped - CodeRabbit’s high-level summary is enabled.
Title check ✅ Passed The title uses a valid Conventional Commit format and accurately describes the documentation changes for NVCF-native MNNVL scheduling primitives.
✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Commit unit tests in branch estroczynski/docs/legacy-nvlink-sched

Comment @coderabbitai help to get the list of available commands.

@estroz estroz changed the title docs(topology): document NVCF dra labels on charts for nvlink scheduling docs(topology): document NVCF-native MNNVL scheduling primitives Aug 10, 2026
@github-actions

Copy link
Copy Markdown
Contributor

@sbaum1994
sbaum1994 added this pull request to the merge queue Aug 10, 2026
Merged via the queue into main with commit 551bd38 Aug 10, 2026
18 checks passed
@sbaum1994
sbaum1994 deleted the estroczynski/docs/legacy-nvlink-sched branch August 10, 2026 19:41
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants