[AIDynamo] comparison report

[podkidyshev]

Summary

Add comparison report for AI Dynamo

Test Plan

• Automated CI
• Manual runs

Additional Notes

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Enterprise

Run ID: 3cff6684-b6c1-4520-b1b9-70804c361386

📥 Commits

Reviewing files that changed from the base of the PR and between 019ea20 and 7757f78.

📒 Files selected for processing (2)

• src/cloudai/workloads/ai_dynamo/report_generation_strategy.py
• tests/workloads/ai_dynamo/test_comparison_report.py

---

📝 Walkthrough

Walkthrough

Changes

AI Dynamo now supports a registered comparison report that parses benchmark and accuracy CSV outputs, normalizes LLM-serving metrics, generates HTML, and is covered by registry and end-to-end tests.

AI Dynamo comparison reporting

Layer / File(s)	Summary
Benchmark CSV normalization src/cloudai/workloads/ai_dynamo/report_generation_strategy.py, src/cloudai/workloads/common/llm_serving_report.py	Parses AI Dynamo metrics into normalized benchmark results, including latency, throughput, request counts, concurrency, GPU usage, accuracy, and missing-value formatting.
Comparison report integration src/cloudai/workloads/ai_dynamo/comparison_report.py, src/cloudai/workloads/ai_dynamo/__init__.py, src/cloudai/registration.py, doc/workloads/ai_dynamo.rst	Adds the AI Dynamo comparison-report adapter, exports its report types, registers the enabled ai_dynamo_comparison scenario report, and documents its metrics and supported CSV outputs.
Registry and end-to-end validation tests/test_init.py, tests/workloads/ai_dynamo/test_comparison_report.py	Validates registry entries and synthetic benchmark/accuracy processing through generated HTML output.

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

Suggested labels: feature

Suggested reviewers: srivatsankrishnan, jeffnvidia

🚥 Pre-merge checks | ✅ 4

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is concise and accurately highlights the main change: adding an AI Dynamo comparison report.
Description check	✅ Passed	The description clearly matches the changeset by stating that it adds a comparison report for AI Dynamo.
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.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch ipod/dynamo-comp

---

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

[coderabbitai]

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 `@src/cloudai/workloads/ai_dynamo/report_generation_strategy.py`:
- Around line 146-150: Update the four percentile assignments in the report
generation strategy to use an explicit None check rather than truthiness when
falling back to unavailable. Apply this to median_ttft_ms, p99_ttft_ms,
median_tpot_ms, and p99_tpot_ms so valid zero values are preserved.

🪄 Autofix (Beta)

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: ASSERTIVE

Plan: Enterprise

Run ID: 5edb70cd-0825-4e9f-9af3-ed3a1109b6df

📥 Commits

Reviewing files that changed from the base of the PR and between c30e75c and 019ea20.

📒 Files selected for processing (8)

• doc/workloads/ai_dynamo.rst
• src/cloudai/registration.py
• src/cloudai/workloads/ai_dynamo/__init__.py
• src/cloudai/workloads/ai_dynamo/comparison_report.py
• src/cloudai/workloads/ai_dynamo/report_generation_strategy.py
• src/cloudai/workloads/common/llm_serving_report.py
• tests/test_init.py
• tests/workloads/ai_dynamo/test_comparison_report.py