Skip to content

Make Directory Name More Unique#60

Merged
michaelmckinsey1 merged 3 commits into
LBANN:mainfrom
michaelmckinsey1:ms-dirname
Mar 3, 2026
Merged

Make Directory Name More Unique#60
michaelmckinsey1 merged 3 commits into
LBANN:mainfrom
michaelmckinsey1:ms-dirname

Conversation

@michaelmckinsey1

@michaelmckinsey1 michaelmckinsey1 commented Jan 29, 2026

Copy link
Copy Markdown
Contributor
  • Addresses Unique naming for torchrun directories #59 by adding the first 8 characters of the UUID to the folder name. Since this stems from jobs running on different processes, this will never conflict in the case of simultaneous jobs.
Previous name:  torchrun_hpc-scaffold_2026-01-29_10h57m26s
New name:       torchrun_hpc-scaffold_2026-01-29_10h57m26s_79ad38c7

@tbennun

tbennun commented Feb 23, 2026

Copy link
Copy Markdown
Collaborator

@michaelmckinsey1 This will still be problematic and asks for the time twice. Instead, what about adding a unique suffix of _{id}?

@michaelmckinsey1

Copy link
Copy Markdown
Contributor Author

@michaelmckinsey1 This will still be problematic and asks for the time twice. Instead, what about adding a unique suffix of _{id}?

This is fine with me. Is the updated solution agreeable?

@michaelmckinsey1 michaelmckinsey1 linked an issue Feb 26, 2026 that may be closed by this pull request

@tbennun tbennun left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

LGTM

@michaelmckinsey1 michaelmckinsey1 merged commit 4a5c36e into LBANN:main Mar 3, 2026
2 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.

Unique naming for torchrun directories

2 participants