luoyuliuyin opened a new issue, #39482:
URL: https://github.com/apache/airflow/issues/39482

   ### Apache Airflow version
   
   2.9.1
   
   ### If "Other Airflow 2 version" selected, which one?
   
   _No response_
   
   ### What happened?
   
   When the scheduler send task to celery, if there is only 1 task in the 
current cycle, the task will be sent to the main thread; if there are multiple 
tasks, a thread pool will be created based on the number of CPU cores, and then 
all tasks will be consumed by the thread pool.
   There are some problems with the current implementation. The scheduler 
creates a thread pool every time it schedules, which will bring a very large 
performance overhead. In fact, the thread pool can be reused.
   
![image](https://github.com/apache/airflow/assets/28948186/266094d9-9d27-4e6b-b549-da6d0884a4b0)
   
   When I tested, sometimes it would take almost 4 seconds to consume 32 tasks.
   
![image](https://github.com/apache/airflow/assets/28948186/62fade6e-61c4-4d29-b147-58b5b6236ac6)
   
   
   ### What you think should happen instead?
   
   Reuse `ProcessPoolExecutor` in `CeleryExecutor`
   
   ### How to reproduce
   
   always
   
   ### Operating System
   
   macOS
   
   ### Versions of Apache Airflow Providers
   
   _No response_
   
   ### Deployment
   
   Official Apache Airflow Helm Chart
   
   ### Deployment details
   
   _No response_
   
   ### Anything else?
   
   _No response_
   
   ### Are you willing to submit PR?
   
   - [X] Yes I am willing to submit a PR!
   
   ### Code of Conduct
   
   - [X] I agree to follow this project's [Code of 
Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md)
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to