Skip to content

Enable prefix caching for RHAIIS CPT pipelines - #275

Merged
Harshith-umesh merged 1 commit into
openshift-psap:mainfrom
Harshith-umesh:codex/rhaiis-cpt-prefix-caching
Sep 28, 2026
Merged

Harshith-umesh merged 1 commit into
openshift-psap:mainfrom
Harshith-umesh:codex/rhaiis-cpt-prefix-caching

Conversation

@Harshith-umesh

@Harshith-umesh Harshith-umesh commented Sep 28, 2026 •

Copy link
Copy Markdown
Member

Summary

Enable prefix caching for every RHAIIS CPT pipeline, using the setting for its runtime:

  • vLLM: rhaiis.engines.vllm.args.enable-prefix-caching: true
  • SGLang: rhaiis.engines.sglang.args.disable-radix-cache: false
  • TensorRT-LLM: rhaiis.engines.trtllm.trtllm_config.kv_cache_config.enable_block_reuse: true

This ensures CPT submissions explicitly enable prefix caching, including where the Forge engine defaults disable it.

Validation

  • git diff --check
  • Automated tests were not run; this is a CPT YAML configuration change.

Summary by CodeRabbit

  • Performance
    • Enabled prefix caching by default in additional vLLM pipelines.
    • Kept radix caching enabled for the SGLang competitive pipeline and enabled KV-cache block reuse for TensorRT-LLM competitive.

@openshift-ci

openshift-ci Bot commented Sep 28, 2026

Copy link
Copy Markdown

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign sjmonson for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Details Needs approval from an approver in each of these files:

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

@coderabbitai

coderabbitai Bot commented Sep 28, 2026 •

Copy link
Copy Markdown

Review in Change Stack →

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info
⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: 19ac6d37-e56e-4b18-a952-d8d3440c9c07

📥 Commits

Reviewing files that changed from the base of the PR and between 4c2472a and 239be42.

📒 Files selected for processing (1)
  • projects/rhaiis/orchestration/cpt.d/cpt.yaml

Included review availability: This review used your included allowance. Your plan provides up to 1 included review per hour; 0 remain after this review.


📝 Walkthrough

Walkthrough

The orchestration configuration sets cache options for vLLM release, competitive, and regression pipelines, and for SGLang and TensorRT-LLM competitive pipelines.

Changes

Pipeline cache settings

Layer / File(s) Summary
vLLM prefix caching
projects/rhaiis/orchestration/cpt.d/cpt.yaml
The vLLM release, competitive, H200 regression, B200 regression, and MI355X regression pipelines enable prefix caching.
Competitive pipeline cache options
projects/rhaiis/orchestration/cpt.d/cpt.yaml
The SGLang competitive pipeline sets disable-radix-cache to false. The TensorRT-LLM competitive pipeline enables KV-cache block reuse.

Priority: ⬇️ Low

Estimated code review effort: 1 (Trivial) | ~4 minutes

Change: Feature

Merge Risk: ⚪ Minimal · up to 239be

No confirmed caching regression blocks merging. The effective TensorRT-LLM setting remains unverified.

Architecture Summary

Architecture risk: 🔵 Low · up to 239be

The change affects 1 system.

Changed systems: projects

Architecture concerns
No architecture-level concerns identified.

Review details

Systems and components

  • observed — projects (service) was modified; 1 changed file maps to changed impact.

Before / after behavior

  • observed — Modified behavior in projects/rhaiis/orchestration/cpt.d/cpt.yaml: The vLLM release pipeline sets enable-prefix-caching to true.
  • observed — Modified behavior in projects/rhaiis/orchestration/cpt.d/cpt.yaml: The SGLang competitive pipeline sets disable-radix-cache to false.
  • observed — Modified behavior in projects/rhaiis/orchestration/cpt.d/cpt.yaml: The TensorRT-LLM competitive pipeline sets KV-cache block reuse to true.
  • observed — Modified behavior in projects/rhaiis/orchestration/cpt.d/cpt.yaml: The vLLM competitive pipeline sets enable-prefix-caching to true.
🚥 Pre-merge checks | ✅ 5
✅ Passed checks (5 passed)
Check name Status Explanation
Description Check ✅ Passed Check skipped - CodeRabbit’s high-level summary is enabled.
Title check ✅ Passed The title clearly and concisely describes the main change: enabling prefix caching for RHAIIS CPT pipelines.
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0…
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 unit tests (beta)
  • Create a new PR

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

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

@Harshith-umesh
Harshith-umesh merged commit e764532 into openshift-psap:main Sep 28, 2026
5 of 6 checks passed
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.

1 participant