Skip to content

Adapt to rapidsmpf async shuffle changes #21787

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
rapids-bot merged 2 commits into from
Mar 16, 2026
Merged

Adapt to rapidsmpf async shuffle changes #21787

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
18c3ea7
Adapt to rapidsmpf async shuffle changes
wence- Mar 13, 2026
e92ae21
Merge branch 'main' into wence/fea/adapt-rapidsmpf-shuffle
wence- Mar 16, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 3 additions & 7 deletions python/cudf_polars/cudf_polars/experimental/rapidsmpf/collectives/shuffle.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,7 @@ async def insert_finished(self) -> None:
"""Insert finished into the ShuffleManager."""
await self.shuffler.insert_finished(self.context)

async def extract_chunk(self, sequence_number: int, stream: Stream) -> plc.Table:
def extract_chunk(self, sequence_number: int, stream: Stream) -> plc.Table:
"""
Extract a chunk from the ShuffleManager.

Expand All @@ -130,11 +130,7 @@ async def extract_chunk(self, sequence_number: int, stream: Stream) -> plc.Table
KeyError
If the requested sequence number has already been extracted.
"""
partition_chunks = await self.shuffler.extract_async(
self.context, sequence_number
)
if partition_chunks is None:
raise KeyError(f"Partition {sequence_number} has already been extracted")
partition_chunks = self.shuffler.extract(sequence_number)
return py_unpack_and_concat(
partitions=partition_chunks,
stream=stream,
Expand Down Expand Up @@ -244,7 +240,7 @@ async def _global_shuffle(
Message(
partition_id,
TableChunk.from_pylibcudf_table(
table=await shuffle.extract_chunk(partition_id, stream),
table=shuffle.extract_chunk(partition_id, stream),
stream=stream,
exclusive_view=True,
),
Expand Down
2 changes: 1 addition & 1 deletion python/cudf_polars/cudf_polars/experimental/rapidsmpf/groupby.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -461,7 +461,7 @@ async def _shuffle_reduce(
for partition_id in shuffle.local_partitions():
stream = ir_context.get_cuda_stream()
partition_chunk = TableChunk.from_pylibcudf_table(
await shuffle.extract_chunk(partition_id, stream),
shuffle.extract_chunk(partition_id, stream),
stream,
exclusive_view=True,
)
Expand Down
Loading