Skip to content

GCSFuse cache enabled a3-mega blueprint#3460

Merged
samskillman merged 5 commits into
GoogleCloudPlatform:developfrom
koallison:gcsfuse_blueprint
Feb 4, 2025
Merged

GCSFuse cache enabled a3-mega blueprint#3460
samskillman merged 5 commits into
GoogleCloudPlatform:developfrom
koallison:gcsfuse_blueprint

Conversation

@koallison
Copy link
Copy Markdown
Contributor

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 #

@tpdownes tpdownes self-assigned this Jan 10, 2025
Copy link
Copy Markdown
Contributor

@tpdownes tpdownes left a comment

Choose a reason for hiding this comment

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

Looks good with some minor suggestions

Comment thread examples/machine-learning/a3-megagpu-8g/slurm-a3mega-gcsfuse-lssd-cluster.yaml Outdated
Comment thread examples/machine-learning/a3-megagpu-8g/slurm-a3mega-gcsfuse-lssd-cluster.yaml Outdated
Comment thread examples/machine-learning/a3-megagpu-8g/slurm-a3mega-gcsfuse-lssd-cluster.yaml Outdated
@tpdownes tpdownes assigned koallison and unassigned tpdownes Jan 10, 2025
@koallison koallison requested a review from tpdownes January 14, 2025 19:34
Copy link
Copy Markdown
Contributor

@tpdownes tpdownes left a comment

Choose a reason for hiding this comment

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

When I make the suggested change, I still can't read the bucket from the controller. What should my expectation be?

root@a3mega-controller:~# ls -lh /gcs
ls: reading directory '/gcs': Operation not supported
total 0
root@a3mega-controller:~# cd /gcs/
root@a3mega-controller:/gcs# ls
ls: reading directory '.': Operation not supported
root@a3mega-controller:/gcs# 

@tpdownes
Copy link
Copy Markdown
Contributor

When I make the suggested change, I still can't read the bucket from the controller. What should my expectation be?

root@a3mega-controller:~# ls -lh /gcs
ls: reading directory '/gcs': Operation not supported
total 0
root@a3mega-controller:~# cd /gcs/
root@a3mega-controller:/gcs# ls
ls: reading directory '.': Operation not supported
root@a3mega-controller:/gcs# 

Offline discussion reveals that this is the default behavior of GCS Fuse. One can read from a specific bucket by suppling /gcs/bucket-name.

@tpdownes
Copy link
Copy Markdown
Contributor

/gcbrun

@tpdownes tpdownes added the release-key-new-features Added to release notes under the "Key New Features" heading. label Jan 15, 2025
@tpdownes tpdownes assigned tpdownes and unassigned koallison Jan 18, 2025
@koallison koallison requested a review from tpdownes January 24, 2025 15:22
@nadig-google nadig-google requested review from nadig-google and removed request for tpdownes February 1, 2025 01:10
@nadig-google nadig-google assigned nadig-google and unassigned tpdownes Feb 1, 2025
Copy link
Copy Markdown
Contributor

@nadig-google nadig-google left a comment

Choose a reason for hiding this comment

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

Approving (Tom is OOO, got approval from Ivan in chat)

@samskillman samskillman dismissed tpdownes’s stale review February 4, 2025 16:39

Requested changes have been addressed.

@samskillman
Copy link
Copy Markdown
Collaborator

/gcbrun

@samskillman samskillman merged commit 46baa48 into GoogleCloudPlatform:develop Feb 4, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

release-key-new-features Added to release notes under the "Key New Features" heading.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants