NVIDIA / NVIDIA/cudf

[FEA] simplify `page_state_s` and possibly other structures for specialized parquet kernels

Open
#15,267 0 comments 0 reactions 0 assignees View on GitHub
cuIO feature request libcudf Performance
Dominant language
C++
Stars
9.8k
Forks
1.1k
Avg merge
3d 6m
Merged PRs (30d)
278

Description

@vuule brought up the point here https://github.com/rapidsai/cudf/pull/15159#discussion_r1505126922 that `page_state_s` could be simplified for the fixed/fixed dictionary kernels, as they don't need all the members of this structure and it might help with shared memory use.

I also think there are other shared structures that we should audit and clean up. I am filing this issue so we don't loose track of it.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.