Release Notes:

1. Multi-Architectural Support:

• The tool is now able to support COS based GCE VM instances with both X86_64 and aarach64 architectures and therefore is able to operate on ARM based GPU VM instances such as a4x-highgpu-4g.
• The Dockerfile (eg. docker build commands) as well as certain code and comments have been updated to reflect this change.

2. Documentation update:

• Provided a more versatile command in the Quick Start section that can dynamically mount the GPU device volumes based on the available devices on the system. This replaced the previously hardcoded command as the number of GPUs in the system varies among VM shapes.

3. Improved the logic on exporting the MFT logs to GCS bucket:

• The tool would to lookup the bucket directly and check its existence instead of creating the bucket directly and using try except to handle the case where the bucket exists. This is expected to improve code health and avoids some edge case failure patterns from GCS bucket creation.

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 #