Support large-file splitting between ranks in read_parquet
#736
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
rjzamora
added
feature request
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
Member
Author
|
/ok to test |
Member
Author
|
/ok to test |
madsbk
reviewed
Dec 16, 2025
cpp/src/streaming/cudf/parquet.cpp
Outdated
| std::back_inserter(chunk_files) | ||
| bool rank_has_assigned_work = false; // Track if this rank was assigned work | ||
|
|
||
| if (files.size() >= size) { |
Member
There was a problem hiding this comment.
I think it would be very helpful to move the two cases into separate helper functions, with docstrings describing the two different approaches.
Member
Author
There was a problem hiding this comment.
Yeah, I definitely think that makes sense. I refactored the chunk-assignment logic into two distinct helper functions: assign_chunks_standard and assign_chunks_split_files.
@wence- - Let me know if this refactor is disruptive to anything you may have in flight.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
Addresses item 2 in #615
General Approach: When the number of files is smaller than the number of ranks, we estimate the number of chunks we need to produce for each file, and use this estimate to determine the (row-group-aligned) boundary between ranks.
This PR does not attempt to optimize metadata collection or to align chunks with row-group boundaries. I'd like that to separate that work for now.