Skip to content

Add opt-in background collation for BaseDistLoader#512

Draft
kmontemayor2-sc wants to merge 3 commits intomainfrom
kmonte/prefetch-data-loaders
Draft

Add opt-in background collation for BaseDistLoader#512
kmontemayor2-sc wants to merge 3 commits intomainfrom
kmonte/prefetch-data-loaders

Conversation

@kmontemayor2-sc
Copy link
Collaborator

Move _collate_fn execution to a daemon thread so it overlaps with GPU training. Controlled by a new background_collation_queue_size parameter (None = disabled, positive int = buffer depth). The thread continuously recv→collate→enqueue while the main thread dequeues pre-collated batches.

Scope of work done

Where is the documentation for this feature?: N/A

Did you add automated tests or write a test plan?

Updated Changelog.md? NO

Ready for code review?: NO

Move _collate_fn execution to a daemon thread so it overlaps with GPU
training. Controlled by a new background_collation_queue_size parameter
(None = disabled, positive int = buffer depth). The thread continuously
recv→collate→enqueue while the main thread dequeues pre-collated batches.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@kmontemayor2-sc
Copy link
Collaborator Author

/all_test

@github-actions
Copy link
Contributor

github-actions bot commented Feb 24, 2026

GiGL Automation

@ 23:02:05UTC : 🔄 E2E Test started.

@ 24:21:44UTC : ✅ Workflow completed successfully.

@github-actions
Copy link
Contributor

github-actions bot commented Feb 24, 2026

GiGL Automation

@ 23:02:06UTC : 🔄 Scala Unit Test started.

@ 23:11:01UTC : ✅ Workflow completed successfully.

@github-actions
Copy link
Contributor

github-actions bot commented Feb 24, 2026

GiGL Automation

@ 23:02:08UTC : 🔄 Integration Test started.

@ 24:26:48UTC : ✅ Workflow completed successfully.

@github-actions
Copy link
Contributor

github-actions bot commented Feb 24, 2026

GiGL Automation

@ 23:02:08UTC : 🔄 Lint Test started.

@ 23:09:05UTC : ✅ Workflow completed successfully.

@github-actions
Copy link
Contributor

github-actions bot commented Feb 24, 2026

GiGL Automation

@ 23:02:10UTC : 🔄 Python Unit Test started.

@ 24:14:14UTC : ✅ Workflow completed successfully.

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.

2 participants