NVIDIA / NVIDIA/cudf

[PERF]: Fully adopt hybrid scan reader in cudf-polars

Open
#23,539 0 comments 0 reactions 0 assignees View on GitHub
cudf-polars Performance
Dominant language
C++
Stars
9.8k
Forks
1.1k
Avg merge
3d 6m
Merged PRs (30d)
278

Description

To fully adopt the hybrid scan reader we need to dispatch to it for both scan types:

- **SplitScan** (single file, row-group aligned splits): partially done in PR #23677
- **FusedScan** (multi-file reads): not yet implemented. Requires Python bindings for the multi-file hybrid scan reader in pylibcudf (PR #22795).

Note: libcudf may eventually auto-dispatch to the correct single or multi-file reader internally.

Contributor guide

Open the contributing guide

Research direction

Start in cudf-polars scan dispatch for SplitScan and FusedScan, then inspect the pylibcudf multi-file hybrid scan reader bindings described in PR #22795. Compare the existing SplitScan work in PR #23677 and verify that both single-file row-group-aligned and multi-file reads dispatch to hybrid readers; done when both scan types are covered.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, python
Domain
data, performance
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.