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

PERF: Use block-wise parallelism for unique. #4916

Open
mvashishtha opened this issue Sep 1, 2022 · 0 comments
Open

PERF: Use block-wise parallelism for unique. #4916

mvashishtha opened this issue Sep 1, 2022 · 0 comments
Labels
P2 Minor bugs or low-priority feature requests Performance 🚀 Performance related issues and pull requests.

Comments

@mvashishtha
Copy link
Collaborator

unique is a full-axis operation in Modin. We should see whether leveraging parallelism across both axes can help us. We could try something like

  1. get unique for each block partition
  2. combine results for each column partition
  3. get unique for each merged result

This came up here: #4798 (comment)

cc @vnlitvinov

@mvashishtha mvashishtha added Performance 🚀 Performance related issues and pull requests. P2 Minor bugs or low-priority feature requests labels Sep 1, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
P2 Minor bugs or low-priority feature requests Performance 🚀 Performance related issues and pull requests.
Projects
None yet
Development

No branches or pull requests

1 participant