Skip to content

Occupancy improvement for distinct hash join with specialized dispatch - #17290

Closed
tgujar wants to merge 3 commits into
NVIDIA:branch-24.12from
tgujar:distinct_hash_join_occupancy
Closed

Occupancy improvement for distinct hash join with specialized dispatch #17290
tgujar wants to merge 3 commits into
NVIDIA:branch-24.12from
tgujar:distinct_hash_join_occupancy

Conversation

@tgujar

@tgujar tgujar commented Nov 9, 2024

Copy link
Copy Markdown
Contributor

Description

MR adds specialized dispatch for distinct hash join insert and probe operations to avoid high occupancy cost of complex data types.
This is a split of the original MR #15700 so that its easier to review and merge. It contains all the changes from previous comments on relevant files.
Addresses #15502 for distinct hash joins

Checklist

  • I am familiar with the Contributing Guidelines.
  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

@tgujar
tgujar requested a review from a team as a code owner November 9, 2024 03:05
@tgujar
tgujar requested review from harrism and shrshi November 9, 2024 03:05
@copy-pr-bot

copy-pr-bot Bot commented Nov 9, 2024

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@github-actions github-actions Bot added the libcudf Affects libcudf (C++/CUDA) code. label Nov 9, 2024
@PointKernel
PointKernel self-requested a review November 9, 2024 03:08
@PointKernel PointKernel added breaking Breaking change improvement Improvement / enhancement to an existing function Performance Performance related issue 3 - Ready for Review Ready for review by team labels Nov 10, 2024
Comment thread cpp/include/cudf/table/experimental/row_operators.cuh
@PointKernel

Copy link
Copy Markdown
Member

/ok to test

@vyasr

vyasr commented May 16, 2025

Copy link
Copy Markdown
Contributor

@PointKernel @tgujar is this PR still needed? I'm not sure how it fits in with #15700 and #17726

@PointKernel

Copy link
Copy Markdown
Member

Closing this PR as it is superseded by #17726

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

3 - Ready for Review Ready for review by team breaking Breaking change improvement Improvement / enhancement to an existing function libcudf Affects libcudf (C++/CUDA) code. Performance Performance related issue

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants