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

Kubernetes Executor wastefully creates pods for DummyOperator tasks #7871

Closed
jpuffer opened this issue Mar 25, 2020 · 4 comments
Closed

Kubernetes Executor wastefully creates pods for DummyOperator tasks #7871

jpuffer opened this issue Mar 25, 2020 · 4 comments
Labels
kind:feature Feature Requests

Comments

@jpuffer
Copy link

jpuffer commented Mar 25, 2020

Description

Kubernetes Executor currently spins up a whole worker pod to execute a dummy task. We know Airflow's DummyOperator doesn't actually do work - it's there to help to organize /group tasks and very useful for that purpose.

Use case / motivation

For complex DAGs, skipping the worker creation process for dummy tasks could significantly improve execution time and resource use in the cluster. Could the executor work more like the local executor when dealing with the dummy tasks? (that way, they could be essentially instant and not interfere with the actual worker pool)

@jpuffer jpuffer added the kind:feature Feature Requests label Mar 25, 2020
@boring-cyborg
Copy link

boring-cyborg bot commented Mar 25, 2020

Thanks for opening your first issue here! Be sure to follow the issue template!

@mik-laj
Copy link
Member

mik-laj commented Mar 26, 2020

Interesting idea. I think you can just mark these tasks automatically in a success state. This made this change work for all executors.

@mik-laj
Copy link
Member

mik-laj commented Mar 26, 2020

Done :-D

@jpuffer
Copy link
Author

jpuffer commented Mar 26, 2020 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
kind:feature Feature Requests
Projects
None yet
Development

No branches or pull requests

2 participants