Change ProcessGroupNCCL default timeout to 10 min #110947
Closed
Conversation
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
Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/110947
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 9aa9a39 with merge base 733368a ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. [ghstack-poisoned]
wconstab
requested review from
H-Huang,
awgu,
d4l3k,
fduwjj,
fegin,
kiukchung,
kwen2501,
mrshenli,
rohan-varma,
wanchaol,
wz337 and
zhaojuanmao
as code owners
Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. [ghstack-poisoned]
wconstab
added a commit
that referenced
this pull request
Oct 10, 2023
Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. ghstack-source-id: 8ab4e9d Pull Request resolved: #110947
fduwjj
reviewed
Oct 10, 2023
|
|
||
| ProcessGroupNCCL::Options::Options(bool is_high_priority_stream) | ||
| : Backend::Options(NCCL_BACKEND_NAME), | ||
| : Backend::Options(NCCL_BACKEND_NAME, kProcessGroupNCCLDefaultTimeout), |
There was a problem hiding this comment.
Why do we need this Options here?
There was a problem hiding this comment.
we seem to have a subclass of Options for every single entity in c10d. idk if it's a good pattern.
fduwjj
reviewed
Oct 10, 2023
Comment on lines
182
to
183
| bool checkDesyncTimeout( | ||
| c10::optional<std::chrono::milliseconds> timeout = c10::nullopt); |
There was a problem hiding this comment.
Is this used somewhere or do you plan to add two-tier timeout here?
There was a problem hiding this comment.
ohh good catch this is not intended to land
This was referenced Oct 10, 2023
Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. [ghstack-poisoned]
Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. [ghstack-poisoned]
wconstab
added a commit
that referenced
this pull request
Oct 10, 2023
Avoid changing default for other backends as CPU backend (GLOO) may need longer timeouts. Motivated by trying to save cluster time when encountering collective hangs. Generally collectives should time out within seconds and 30 minutes (or 10 minutes) should provide ample headroom for edge cases. ghstack-source-id: 6aad49d Pull Request resolved: #110947
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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.
Stack from ghstack (oldest at bottom):
Avoid changing default for other backends as CPU backend (GLOO) may need
longer timeouts.
Motivated by trying to save cluster time when encountering collective
hangs. Generally collectives should time out within seconds and 30
minutes (or 10 minutes) should provide ample headroom for edge cases.