Skip to content

Update Dockerfile and README.md instructions for a3mega nemo framework#5164

Merged
mufaqam-gcl merged 4 commits intoGoogleCloudPlatform:developfrom
mufaqam-gcl:fix/a3mega-nemo
Feb 2, 2026
Merged

Update Dockerfile and README.md instructions for a3mega nemo framework#5164
mufaqam-gcl merged 4 commits intoGoogleCloudPlatform:developfrom
mufaqam-gcl:fix/a3mega-nemo

Conversation

@mufaqam-gcl
Copy link
Contributor

The Dockerfile needed update to the env variables and the instructions in README.md also needed to be updated.

Submission Checklist

NOTE: Community submissions can take up to 2 weeks to be reviewed.

Please take the following actions before submitting this pull request.

  • Fork your PR branch from the Toolkit "develop" branch (not main)
  • Test all changes with pre-commit in a local branch #
  • Confirm that "make tests" passes all tests
  • Add or modify unit tests to cover code changes
  • Ensure that unit test coverage remains above 80%
  • Update all applicable documentation
  • Follow Cluster Toolkit Contribution guidelines #

@mufaqam-gcl mufaqam-gcl requested review from a team and samskillman as code owners February 1, 2026 07:59
@mufaqam-gcl mufaqam-gcl added the release-improvements Added to release notes under the "Improvements" heading. label Feb 1, 2026
@mufaqam-gcl mufaqam-gcl changed the base branch from main to develop February 1, 2026 08:00
@mufaqam-gcl mufaqam-gcl enabled auto-merge (squash) February 1, 2026 08:00
@gemini-code-assist
Copy link
Contributor

Summary of Changes

Hello @mufaqam-gcl, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request enhances the framework's robustness and usability by updating core dependencies, improving input validation mechanisms across various modules, and expanding support for advanced GKE TPU configurations. It also refines GCSFuse integration for machine learning workloads and optimizes the continuous integration pipeline with new tests and build process simplifications.

Highlights

  • Dockerfile and README Updates: The Dockerfile for the a3mega nemo framework has been updated with revised NCCL environment variables, and the corresponding README.md instructions have been adjusted to reflect these changes, including new data download steps and container mount paths.
  • Dependency Updates: Numerous Python package dependencies in community/front-end/ofe/requirements.txt have been updated to their latest versions, and Go dependencies (golang.org/x/oauth2, golang.org/x/sys) have also been bumped.
  • New Range Validator for Metadata: A new RangeValidator has been implemented and integrated into the blueprint validation system. This validator allows defining minimum and maximum bounds for numeric input variables or the length of lists in metadata.yaml files, enhancing input validation capabilities across various modules.
  • Slurm Cluster Name Regex Expansion: The regular expression for slurm_cluster_name validation in several Slurm-related modules (gke-nodeset, schedmd-slurm-gcp-v6-controller, slurm-gcp/instance_template, slurm-gcp/internal_instance_template) has been updated to allow hyphens and extend the maximum length from 10 to 20 characters, providing more flexibility for cluster naming.
  • Shared Reservation Permission Clarification: Documentation and code related to shared reservations in schedmd-slurm-gcp-v6-nodeset have been updated to clarify that while the compute.reservations.get permission is ideal, the system will now gracefully fall back to minimal reservation settings and proceed if this permission is missing.
  • GKE TPU Flex Start Examples: New examples and configurations have been added for GKE TPU Flex Start clusters, specifically for TPU 7x and TPU v6e, including deployment YAMLs, Kueue configurations, and detailed READMEs on how to set up and test dynamic scaling for these TPU types.
  • GCSFuse Mount Options Refinement: GCSFuse mount options in several machine learning blueprints (a3-ultragpu-8g, a4-highgpu-8g, a4x-highgpu-4g) have been updated to use underscore-separated keys and leverage new profile options (e.g., aiml-checkpointing, aiml-training, aiml-serving) for optimized performance based on workload type.
  • Cloud Build Test Infrastructure Improvements: Cloud Build test configurations have been streamlined by removing redundant gcluster-bundle.zip download/unzip steps and associated secretEnv variables, replacing them with a local make command. New daily tests for GKE H4D on-spot and ML H4D on-spot Slurm have also been introduced.
Ignored Files
  • Ignored by pattern: .github/workflows/** (1)
    • .github/workflows/pr-precommit.yml
Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

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

Code Review

This pull request updates environment variables in the Dockerfile for the NeMo framework on A3 MegaGPUs and enhances the setup instructions in the README. The changes seem geared towards performance tuning and improving the user experience for this example. My review includes one suggestion for the README to use more stable, canonical URLs for downloading model assets, which will improve long-term maintainability.

@mufaqam-gcl mufaqam-gcl mentioned this pull request Feb 2, 2026
Copy link
Contributor

@arpit974 arpit974 left a comment

Choose a reason for hiding this comment

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

Could you share any benchmark results if they've been run? It would be great to see the effect of these optimizations.

@mufaqam-gcl mufaqam-gcl merged commit 323e370 into GoogleCloudPlatform:develop Feb 2, 2026
12 of 75 checks passed
AdarshK15 pushed a commit to AdarshK15/cluster-toolkit that referenced this pull request Feb 5, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

release-improvements Added to release notes under the "Improvements" heading.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants