Skip to content

docs(planner): add power-aware design - #12701

Closed
kaim-eng wants to merge 7 commits into
mainfrom
docs/power-planner-design-v2
Closed

docs(planner): add power-aware design#12701
kaim-eng wants to merge 7 commits into
mainfrom
docs/power-planner-design-v2

Conversation

@kaim-eng

@kaim-eng kaim-eng commented Aug 5, 2026

Copy link
Copy Markdown
Contributor

Summary

  • Add a Developer Guide design page for the static GPU power-aware Planner behavior implemented in feat(planner): DGD-owned GPU power caps and budget-aware scaling #12012.
  • Document power-cap ownership, admission validation, startup settlement, rollout safety, budget enforcement, metrics, tests, and current guarantees.
  • Record dynamic power-control work as deferred design scope and add the page to the Planner navigation.

Validation

  • git diff --check
  • pre-commit run --files docs/fern/pages/developer-guide/knowledge-base/modular-components/planner/power-aware-planner-design.md
    • All applicable content checks passed.
    • The Windows host could not run check-asset-paths because Python selected CP1252 for UTF-8 docs, or pytest-marker-report because it imports Linux-only fcntl.
  • WSL: python3 docs/fern/scripts/check_asset_paths.py (389 files checked)
  • WSL: python3 docs/fern/scripts/convert_callouts.py --test (19 tests passed)

Reviewer Guidance

Start with:

  • Decision Summary for the current static contract.
  • Ownership and Data Flow and Admission Contract for component boundaries.
  • Deferred Dynamic Control and Follow-Up Priorities for the intentionally deferred design scope.

The implementation is already merged in #12012. This PR documents that behavior; it does not add runtime functionality.

Related Issues

Signed-off-by: Kai Ma <kaim@nvidia.com>
@kaim-eng
kaim-eng requested a review from a team as a code owner August 5, 2026 14:58
@copy-pr-bot

copy-pr-bot Bot commented Aug 5, 2026

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@github-actions github-actions Bot added docs documentation Improvements or additions to documentation labels Aug 5, 2026
@coderabbitai

coderabbitai Bot commented Aug 5, 2026

Copy link
Copy Markdown
Contributor

Review Change Stack

Walkthrough

The PR adds a Developer Guide page for static GPU power-aware Planner behavior. It documents admission, configuration, startup settlement, rollout handling, budget enforcement, metrics, tests, deferred work, and implementation references.

Changes

Power-Aware Planner

Layer / File(s) Summary
Contracts, ownership, and configuration
docs/fern/index.yml, docs/fern/pages/developer-guide/knowledge-base/modular-components/planner/power-aware-planner-design.md
Adds the Planner navigation entry and documents DGD admission, cap ownership, Pod access, configuration, and static power data flow.
Startup settlement and rollout state
docs/fern/pages/developer-guide/knowledge-base/modular-components/planner/power-aware-planner-design.md
Documents annotation settlement, cached power projections, immutable state, terminating-Pod handling, and rollout scale-up safety.
Proposal budgeting and validation
docs/fern/pages/developer-guide/knowledge-base/modular-components/planner/power-aware-planner-design.md
Defines proposal processing, ordered GPU and power clamps, guarantees, metrics, and test coverage.
Deferred work and implementation map
docs/fern/pages/developer-guide/knowledge-base/modular-components/planner/power-aware-planner-design.md
Lists deferred design areas, implementation references, associated tests, and related documentation. לאחר

Estimated code review effort: 3 (Moderate) | ~25 minutes

🚥 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.
Title check ✅ Passed The title clearly identifies the documentation change and the planner topic.
Description check ✅ Passed The description explains the documentation scope, validation, reviewer starting points, and confirms that no issue is linked.

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

@devin-ai-integration devin-ai-integration Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Devin Review found 1 potential issue.

Open in Devin Review

Signed-off-by: Kai Ma <kaim@nvidia.com>

@coderabbitai coderabbitai Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In
`@docs/fern/pages/developer-guide/knowledge-base/modular-components/planner/power-aware-planner-design.md`:
- Around line 8-14: Update the Tier 3 and Experimental notices in the planner
design documentation to use GitHub-style admonition syntax, using appropriate
alert types and keeping each alert separated by a blank line. Preserve the
existing notice text and links.
🪄 Autofix

Fix all unresolved CodeRabbit comments on this PR:

  • Push a commit to this branch (recommended)
  • Create a new PR with the fixes

ℹ️ Review info
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 87cb6a39-9b3d-4dca-9947-c1ba662e90d6

📥 Commits

Reviewing files that changed from the base of the PR and between 64d6169 and d4ed1d4.

📒 Files selected for processing (2)
  • docs/fern/index.yml
  • docs/fern/pages/developer-guide/knowledge-base/modular-components/planner/power-aware-planner-design.md

@kaim-eng

kaim-eng commented Aug 5, 2026

Copy link
Copy Markdown
Contributor Author

/ok to test d4ed1d4

@github-actions

github-actions Bot commented Aug 5, 2026

Copy link
Copy Markdown
Contributor

Signed-off-by: Kai Ma <kaim@nvidia.com>
Signed-off-by: Kai Ma <kaim@nvidia.com>
Signed-off-by: Kai Ma <kaim@nvidia.com>
…sign-v2

Signed-off-by: Kai Ma <kaim@nvidia.com>
@kaim-eng

Copy link
Copy Markdown
Contributor Author

/ok to test cf34b15

Signed-off-by: Kai Ma <kaim@nvidia.com>

Copy link
Copy Markdown
Contributor Author

Superseded by #13454. The newer PR preserves the complete power-aware Planner design and Fern navigation changes from this PR, and extends them with the Phase 2 runtime qualification contract, exact GPU product/qualification-catalog boundary, product canary procedure, and DGDR follow-up. Closing this PR in favor of #13454.

@kaim-eng kaim-eng closed this Aug 18, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

docs documentation Improvements or additions to documentation size/XL

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant