Multi-rank fixes for cudf-polars streaming - #22361
Merged
Merged
Conversation
github-actions
Bot
added
Python
madsbk
marked this pull request as ready for review
May 4, 2026 10:11
Matt711
requested changes
May 4, 2026
Matt711
left a comment
Matt711
left a comment
Member
There was a problem hiding this comment.
Mostly looks good. We should open an issue for the concat ordering issue I think.
Matt711
approved these changes
May 5, 2026
wence-
approved these changes
May 5, 2026
Contributor
Author
|
/merge |
3 tasks
galipremsagar
pushed a commit
to galipremsagar/cudf
that referenced
this pull request
May 6, 2026
Fix bugs that appear when running with `num_ranks > 1`, where client-side `pl.concat(per_rank_outputs)` exposes assumptions that do not hold under single-rank execution. These were all discovered while working on multi-rank tests. **NB:** Please take a close look during review, as I’m still a bit unfamiliar with the IR part of cudf-polars. Authors: - Mads R. B. Kristensen (https://github.com/madsbk) Approvers: - Matthew Murray (https://github.com/Matt711) - Lawrence Mitchell (https://github.com/wence-) URL: NVIDIA#22361
shrshi
pushed a commit
to shrshi/cudf
that referenced
this pull request
May 12, 2026
Fix bugs that appear when running with `num_ranks > 1`, where client-side `pl.concat(per_rank_outputs)` exposes assumptions that do not hold under single-rank execution. These were all discovered while working on multi-rank tests. **NB:** Please take a close look during review, as I’m still a bit unfamiliar with the IR part of cudf-polars. Authors: - Mads R. B. Kristensen (https://github.com/madsbk) Approvers: - Matthew Murray (https://github.com/Matt711) - Lawrence Mitchell (https://github.com/wence-) URL: NVIDIA#22361
rapids-bot Bot
pushed a commit
that referenced
this pull request
May 13, 2026
- Reverts #22361, which suppressed replicated chunks at intermediate operators and broke downstream consumers under multi-rank. - Dedupe replicated outputs once, at the Dask and Ray frontends where the client concatenates per-rank results. SPMD returns per-rank results unchanged. - `ChannelMetadata.duplicated=True` now consistently means every rank holds the data. - Depends on #22381 Authors: - Matthew Murray (https://github.com/Matt711) Approvers: - Lawrence Mitchell (https://github.com/wence-) URL: #22394
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
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.
Fix bugs that appear when running with
num_ranks > 1, where client-sidepl.concat(per_rank_outputs)exposes assumptions that do not hold under single-rank execution.These were all discovered while working on multi-rank tests.
NB: Please take a close look during review, as I’m still a bit unfamiliar with the IR part of cudf-polars.