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

[BUG] Shuffle partitions for the first GPU run should not be set for more than 200 #1400

Closed
kuhushukla opened this issue Oct 29, 2024 · 1 comment · Fixed by #1479
Closed
Assignees
Labels
bug Something isn't working core_tools Scope the core module (scala)

Comments

@kuhushukla
Copy link
Collaborator

Describe the bug
In order to fully utilize the benefits of multithreaded shuffle, the first GPU run (when autotuner runs on cpu logs) should not increase partitions to values > 200.

We might have to fork this logic for GPU iterations where the benefit of higher partitions overrides the ones from MT shuffle.

@kuhushukla kuhushukla added ? - Needs Triage bug Something isn't working labels Oct 29, 2024
@amahussein amahussein added the core_tools Scope the core module (scala) label Nov 1, 2024
@mattahrens
Copy link
Collaborator

Scope: update auto-tuner to detect that it is processing a CPU event log and then default shuffle partitions to 200. When GPU event log is detected, it can keep existing logic.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working core_tools Scope the core module (scala)
Projects
None yet
4 participants