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

Use parse_url kernel for PATH parsing #10740

Merged
merged 2 commits into from
Apr 29, 2024

Conversation

thirtiseven
Copy link
Collaborator

@thirtiseven thirtiseven commented Apr 25, 2024

This PR adds plugin support for PATH parsing in parse_url.

Contributes to #8963

JNI PR (merged) : NVIDIA/spark-rapids-jni#1997

All cases from the kaggle dataset and current integration test passed.

Perf test results

dataset: kaggle dataset repeated 10 times.

spark.time(df.selectExpr("COUNT(parse_url(url, 'PATH')) as pr1", "COUNT(parse_url(url, 'PATH')) as pr2", "COUNT(parse_url(url, 'PATH')) as pr3", "COUNT(parse_url(url, 'PATH')) as pr4", "COUNT(parse_url(url, 'PATH')) as pr5", "COUNT(parse_url(url, 'PATH')) as pr6", "COUNT(parse_url(url, 'PATH')) as pr7", "COUNT(parse_url(url, 'PATH')) as pr8", "COUNT(parse_url(url, 'PATH')) as pr9", "COUNT(parse_url(url, 'PATH')) as pr0").show())
GPU Time (ms) CPU Time (ms) Speed up
1,288.5 35,330 27.42x

Signed-off-by: Haoyang Li <haoyangl@nvidia.com>
@thirtiseven thirtiseven self-assigned this Apr 25, 2024
@thirtiseven thirtiseven marked this pull request as ready for review April 25, 2024 14:42
Signed-off-by: Haoyang Li <haoyangl@nvidia.com>
@thirtiseven
Copy link
Collaborator Author

build

@thirtiseven thirtiseven merged commit df3f0af into NVIDIA:branch-24.06 Apr 29, 2024
44 checks passed
@thirtiseven thirtiseven deleted the parse_uri_path branch April 29, 2024 15:10
@sameerz sameerz added the feature request New feature or request label Apr 29, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
feature request New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants