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

[qob] Read results files 20x faster by not using Scala parallel collections #12854

Merged
merged 2 commits into from
Apr 7, 2023

Conversation

danking
Copy link
Contributor

@danking danking commented Apr 7, 2023

CHANGELOG: In Query-on-Batch, simple pipelines with large numbers of partitions should be substantially faster.

🤦

…ctions

CHANGELOG: In Query-on-Batch, simple pipelines with large numbers of partitions should be substantially faster.

:facepalm:
@danking
Copy link
Contributor Author

danking commented Apr 7, 2023

readAllBytes, Java concurrency, no semaphore:

3.97912865 s.
502.6226030666287 result/s.p
0.007669412278238353 MiB/s.
https://batch.hail.is/batches/7228525/jobs/1

readAllBytes, Java concurrency, semaphore 100

4.931811728 s.
405.53048459760674 result/s.
0.0061879041228882865 MiB/s.
https://batch.hail.is/batches/7228527/jobs/1

readAllBytes, Scala concurrency, semaphore 100

112.306885221 s.
17.808347155780833 result/s.
2.7173381280183156E-4 MiB/s.
https://batch.hail.is/batches/7228528/jobs/1

@danking
Copy link
Contributor Author

danking commented Apr 7, 2023

this commit

3.436348464 s.
582.0131517372215 result/s.
0.008880815913959068 MiB/s.
https://batch.hail.is/batches/7228536/jobs/1

Copy link
Contributor

@tpoterba tpoterba left a comment

Choose a reason for hiding this comment

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

deep depression

@danking danking mentioned this pull request Apr 7, 2023
@danking danking merged commit c65098a into hail-is:main Apr 7, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants