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.

Sign up for GitHub

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

Jump to bottom

Optimize random walks #3460

Merged
merged 12 commits into from
Apr 18, 2023
Merged

Optimize random walks #3460

merged 12 commits into from
Apr 18, 2023

Conversation

jnke2016
Copy link
Contributor

@jnke2016 jnke2016 commented Apr 8, 2023 •
edited
Loading

Our current implementation of random walks exhibits poor performance at the python layer when converting the results to coalesced paths.

This PR

  1. Optimizes the conversion by leveraging cudf.
  2. Adds additional tests to ensure the accuracy of the conversion.
  3. Provide a better description of the API by updating the docstrings.
  4. Fixes the bug when setting legacy_result_type to False

Datasets

  • Number of vertices: 1244330
  • Number of edges: 33554432

Performance comparison before and after the optimization
image

Performance comparison between coalesced(optimized) and padded results
image

closes #3405

@jnke2016 jnke2016 requested a review from a team as a code owner April 8, 2023 17:52
@alexbarghi-nv alexbarghi-nv self-assigned this Apr 11, 2023
@alexbarghi-nv alexbarghi-nv added improvement Improvement / enhancement to an existing function non-breaking Non-breaking change labels Apr 11, 2023
@alexbarghi-nv alexbarghi-nv added this to the 23.06 milestone Apr 11, 2023
@alexbarghi-nv
Copy link
Member

Great work on this PR, just one question @jnke2016 : are you going to apply these changes to the MG version as well?

rlratzel
rlratzel requested changes Apr 13, 2023
View reviewed changes
python/cugraph/cugraph/sampling/random_walks.py Show resolved Hide resolved
python/cugraph/cugraph/sampling/random_walks.py Outdated Show resolved Hide resolved
python/cugraph/cugraph/sampling/random_walks.py Outdated Show resolved Hide resolved
python/cugraph/cugraph/sampling/random_walks.py Outdated Show resolved Hide resolved
python/cugraph/cugraph/sampling/random_walks.py Outdated Show resolved Hide resolved
python/cugraph/cugraph/sampling/random_walks.py Outdated Show resolved Hide resolved
python/cugraph/cugraph/tests/sampling/test_random_walks.py Show resolved Hide resolved
alexbarghi-nv
alexbarghi-nv approved these changes Apr 18, 2023
View reviewed changes
Copy link
Member

@alexbarghi-nv alexbarghi-nv left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

👍

@rlratzel
Copy link
Contributor

/merge

@rapids-bot rapids-bot bot merged commit e23f587 into rapidsai:branch-23.06 Apr 18, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@alexbarghi-nv alexbarghi-nv alexbarghi-nv approved these changes

@rlratzel rlratzel rlratzel approved these changes

Assignees

@jnke2016 jnke2016

Labels
improvement Improvement / enhancement to an existing function non-breaking Non-breaking change
Projects
None yet
Milestone
23.06
Development

Successfully merging this pull request may close these issues.

[BUG]: low utilization and bad performance for GPU random walk on python API
3 participants
@jnke2016 @alexbarghi-nv @rlratzel