Repository navigation
[GLUTEN-10683][VL] Add lock to restrict single task for one GPU - #10684
Merged
Merged
Conversation
jinchengchenghh
force-pushed
the
gpu_task
branch
from
September 11, 2025 14:26
8aa799a to
6d1d48e
Compare
jinchengchenghh
marked this pull request as draft
September 11, 2025 15:59
jinchengchenghh
force-pushed
the
gpu_task
branch
from
September 16, 2025 09:38
a55bf0a to
ba04c36
Compare
jinchengchenghh
marked this pull request as ready for review
September 16, 2025 09:38
jinchengchenghh
force-pushed
the
gpu_task
branch
from
September 16, 2025 14:30
4573585 to
0da5895
Compare
jinchengchenghh
force-pushed
the
gpu_task
branch
from
September 17, 2025 10:32
f878efc to
ceb3209
Compare
zhouyuan
approved these changes
Sep 18, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR facebookincubator/velox#13341 has not been merged, so bypass the operator validation, after it is merged, I will add back.
For GPU task, usually GPU launch one kernel for one GPU, and Spark workload is big dataset, the GPU memory is not enough for parallel task.
With this restriction
✅ TPC-DS Q95 Runtime (in seconds) — 8 Cores (T4 / g4dn.2xlarge)