Make Managed lustre default in A3u and A3m series Slurm blueprints by saara-tyagi27 · Pull Request #5396 · GoogleCloudPlatform/cluster-toolkit

saara-tyagi27 · 2026-03-25T06:57:32Z

Summary

This PR replaces Google Cloud Filestore with Managed Lustre as the default shared filesystem in a3m, a3u and a4h machine learning blueprints

Key Changes:

Storage Migration: Swapped the filestore module for managed-lustre and added the necessary private-service-access networking module.
Default Configuration: Enabled Managed Lustre installation by default ("install_managed_lustre": true).
Test Updates: Updated daily Cloud Build test tags to reflect the filesystem change.

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 #

gemini-code-assist · 2026-03-25T06:57:48Z

Summary of Changes

Hello, 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 transitions the default shared file system for the A3 UltraGPU and A4 HighGPU SLURM blueprints from Filestore to Managed Lustre. This change aims to leverage the performance benefits of Managed Lustre for high-performance computing workloads by updating the blueprint configurations to enable and utilize the new file system, ensuring a more optimized storage solution for these environments.

Highlights

Managed Lustre Integration: The A3 UltraGPU and A4 HighGPU SLURM blueprints have been updated to use Managed Lustre as the default shared file system for /home, replacing Filestore.
Configuration Updates: Variables related to Filestore IP ranges were removed, and Managed Lustre specific variables (instance ID, size, throughput) were enabled and configured in the A4 HighGPU blueprint.
Module Switching: The homefs module in both blueprints was switched from modules/file-system/filestore to modules/file-system/managed-lustre, and the private_service_access module was explicitly enabled.

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.

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. ↩

gemini-code-assist

Code Review

This pull request updates the a3ultra-slurm-blueprint.yaml and a4high-slurm-blueprint.yaml files to transition from using Filestore for /home to Managed Lustre. This involves removing the filestore_ip_range variable, setting install_managed_lustre to true, switching the homefs module source to managed-lustre, and configuring its settings. For a4high-slurm-blueprint.yaml, the Managed Lustre-related variables (lustre_instance_id, lustre_size_gib, per_unit_storage_throughput) are uncommented and initialized. Feedback from the review indicates that in a3ultra-slurm-blueprint.yaml, the Managed Lustre variables are still commented out, which will cause the blueprint to fail. Additionally, obsolete comments related to filestore_ip_range and Managed Lustre instructions in a4high-slurm-blueprint.yaml should be removed to improve clarity and maintainability.

saara-tyagi27 · 2026-03-25T10:33:45Z

/gemini review

gemini-code-assist · 2026-03-25T10:38:13Z

Warning

Gemini encountered an error creating the review. You can try again by commenting /gemini review.

saara-tyagi27 · 2026-03-25T10:43:21Z

/gemini review

gemini-code-assist

Code Review

This pull request updates several machine learning blueprints (A3 HighGPU, A3 MegaGPU, A3 UltraGPU, A4 HighGPU, A4x HighGPU) to replace Filestore with Managed Lustre for shared home directories. The changes involve updating blueprint variables, switching the file system module, adding private service access, and enabling Managed Lustre installation. A consistent improvement opportunity was identified across all updated blueprints: the lustre_instance_id is currently a static string, which could cause conflicts upon multiple deployments. It is recommended to incorporate the deployment_name variable to ensure unique instance IDs.

saara-tyagi27 · 2026-03-31T06:23:26Z

/gcbrun

saara-tyagi27 · 2026-03-31T10:40:48Z

/gemini review

gemini-code-assist

Code Review

This pull request replaces Filestore with Managed Lustre across several A3 and A4 Slurm blueprints, incorporating the private service access module and updating relevant configuration variables and test tags. The review feedback identifies opportunities to improve maintainability and consistency by refining descriptive comments and ensuring uniform quoting of keys within Slurm configuration blocks.

saara-tyagi27 · 2026-04-08T10:39:30Z

/gcbrun

arpit974

please review the failing test and make sure it is not due to this PR change.

saara-tyagi27 · 2026-05-25T06:40:55Z

A3m and A3u tests passed.
slurm-gcp-v6-static - failing, unrelated to my changes in the pr.

saara-tyagi27 force-pushed the filesystem-a-series branch from 958b61f to dff3234 Compare March 25, 2026 06:58

gemini-code-assist Bot reviewed Mar 25, 2026

View reviewed changes

Comment thread examples/machine-learning/a3-ultragpu-8g/a3ultra-slurm-blueprint.yaml

Comment thread examples/machine-learning/a4-highgpu-8g/a4high-slurm-blueprint.yaml

Comment thread examples/machine-learning/a4-highgpu-8g/a4high-slurm-blueprint.yaml Outdated

saara-tyagi27 changed the title ~~use lustre in a3u a4h~~ Make lustre default in A series blueprints Mar 25, 2026

saara-tyagi27 force-pushed the filesystem-a-series branch from dc224fa to f46cbc0 Compare March 25, 2026 10:32

gemini-code-assist Bot reviewed Mar 25, 2026

View reviewed changes

gemini-code-assist Bot reviewed Mar 31, 2026

View reviewed changes

Comment thread examples/machine-learning/a3-highgpu-8g/a3high-slurm-blueprint.yaml Outdated

Comment thread examples/machine-learning/a3-megagpu-8g/a3mega-slurm-blueprint.yaml Outdated

Comment thread examples/machine-learning/a4x-highgpu-4g/a4xhigh-slurm-blueprint.yaml Outdated

saara-tyagi27 force-pushed the filesystem-a-series branch from 1bb5c11 to a36240b Compare March 31, 2026 17:25

saara-tyagi27 changed the title ~~Make lustre default in A series blueprints~~ Make Managed lustre default in A series Slurm blueprints Apr 1, 2026

saara-tyagi27 force-pushed the filesystem-a-series branch 2 times, most recently from c669440 to 4997920 Compare April 7, 2026 03:37

saara-tyagi27 changed the title ~~Make Managed lustre default in A series Slurm blueprints~~ Make Managed lustre default in A3u, A4h, A3m series Slurm blueprints Apr 8, 2026

saara-tyagi27 added enhancement New feature or request release-improvements Added to release notes under the "Improvements" heading. labels Apr 8, 2026

saara-tyagi27 marked this pull request as ready for review April 8, 2026 06:12

saara-tyagi27 requested review from a team and samskillman as code owners April 8, 2026 06:12

saara-tyagi27 force-pushed the filesystem-a-series branch from e890859 to 84fa037 Compare April 9, 2026 05:37

saara-tyagi27 force-pushed the filesystem-a-series branch 2 times, most recently from e27b4bb to 6596542 Compare May 7, 2026 06:21

saara-tyagi27 added 4 commits May 18, 2026 05:48

use lustre in a3u a4h

677c8f7

uncomment variables

0bd7545

Add maanaged lustre and add tags in tests

8565ee0

update lustre_instance_id for deployment uniqueness

179ce06

saara-tyagi27 added 6 commits May 18, 2026 05:48

nit

6b76fc7

Add homefs to slurm_login

d98de6d

Remove a3 high changes

5265a7b

remove a4xhigh changes

68e8bcd

add Managed Lustre to primary VPC naming logic

ec3e2d5

remove maanged lustre changes for a4h

08942ef

saara-tyagi27 force-pushed the filesystem-a-series branch from 6596542 to 08942ef Compare May 20, 2026 07:30

saara-tyagi27 added 2 commits May 20, 2026 10:08

nit

4f2cba1

Add filestore tag back to a4h build file

9570379

saara-tyagi27 changed the title ~~Make Managed lustre default in A3u, A4h, A3m series Slurm blueprints~~ Make Managed lustre default in A3u and A3m series Slurm blueprints May 20, 2026

arpit974 approved these changes May 22, 2026

View reviewed changes

saara-tyagi27 merged commit 493517c into GoogleCloudPlatform:develop May 25, 2026
18 of 83 checks passed

Conversation

saara-tyagi27 commented Mar 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Key Changes:

Submission Checklist

Uh oh!

gemini-code-assist Bot commented Mar 25, 2026

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

saara-tyagi27 commented Mar 25, 2026

Uh oh!

gemini-code-assist Bot commented Mar 25, 2026

Uh oh!

saara-tyagi27 commented Mar 25, 2026

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

saara-tyagi27 commented Mar 31, 2026

Uh oh!

saara-tyagi27 commented Mar 31, 2026

Uh oh!

gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

saara-tyagi27 commented Apr 8, 2026

Uh oh!

arpit974 left a comment

Choose a reason for hiding this comment

Uh oh!

saara-tyagi27 commented May 25, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

saara-tyagi27 commented Mar 25, 2026 •

edited

Loading