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

[data] [streaming] Add autoscaling for ActorPoolTaskSubmitter #31723

Closed
ericl opened this issue Jan 18, 2023 · 0 comments · Fixed by #31987
Closed

[data] [streaming] Add autoscaling for ActorPoolTaskSubmitter #31723

ericl opened this issue Jan 18, 2023 · 0 comments · Fixed by #31987
Assignees
@clarkzinzow
Labels
data Ray Data-related issues enhancement Request for new feature and/or capability P1 Issue that should be fixed within a few weeks
Milestone
Dataset Streaming...

Comments

@ericl
Copy link
Contributor

ericl commented Jan 18, 2023

Enable the actor pool task submitter to autoscale and implement max_tasks_in_flight_per_actor. We should autoscale up when the number of input tasks exceeds that allowed by max_tasks_in_flight, and autoscale down when actors are idle.

cc @clarkzinzow
@ericl ericl added enhancement Request for new feature and/or capability data Ray Data-related issues Ray 2.3 labels Jan 18, 2023
@ericl ericl added the P1 Issue that should be fixed within a few weeks label Jan 18, 2023
This was referenced Jan 19, 2023
Merged
Merged
@jianoaix jianoaix mentioned this issue Jan 25, 2023
Merged
7 tasks
amogkam pushed a commit that referenced this issue Jan 25, 2023
@jianoaix
[Data] Set the num actor to 10 for xgboost batch prediction benchmark (…
fe65c3e 
…#31914)

Autoscaling actor pool is not supported in new execution backend yet: #31723
We temporarily set the actor pool size to 10 (same as the num workers) to unbreak the tests.

Signed-off-by: jianoaix <iamjianxiao@gmail.com>
@clarkzinzow clarkzinzow mentioned this issue Jan 27, 2023
Merged
7 tasks
@jianoaix jianoaix mentioned this issue Feb 2, 2023
Merged
7 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@clarkzinzow clarkzinzow

Labels
data Ray Data-related issues enhancement Request for new feature and/or capability P1 Issue that should be fixed within a few weeks
Projects
None yet
Milestone
Dataset Streaming Execution
2 participants
@ericl @clarkzinzow