Skip to content
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.

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

Improve minhash performance by using more working memory #16756

Draft
wants to merge 28 commits into
base: branch-24.12
Choose a base branch
from

Conversation

davidwendt
Copy link
Contributor

Description

Experimental performance improvement that removes the atomic calls by creating a large working memory to hold intermediate hashes that are then reduced to the output.
The working memory is approximately = (number of input rows x number of seeds x 32) x (size-of hash-type) bytes.
The size-of hash-type would be either 4 bytes (32-bit hash version) or 8 bytes (64-bit hash version).

Checklist

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

@davidwendt davidwendt added 2 - In Progress Currently a work in progress libcudf Affects libcudf (C++/CUDA) code. improvement Improvement / enhancement to an existing function non-breaking Non-breaking change labels Sep 5, 2024
@davidwendt davidwendt self-assigned this Sep 5, 2024
@davidwendt davidwendt changed the base branch from branch-24.10 to branch-24.12 September 23, 2024 23:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
2 - In Progress Currently a work in progress improvement Improvement / enhancement to an existing function libcudf Affects libcudf (C++/CUDA) code. non-breaking Non-breaking change
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant