KevinYang21 commented on issue #5615: [AIRFLOW-5035] Use events/messages not multiprocessing.Manager URL: https://github.com/apache/airflow/pull/5615#issuecomment-514762752 @ashb The graph was from a stress testing cluster we spun up. It is going to be pretty expensive, money and time wise, to spin that up again but I think it shouldn't be too hard to benchmark this change using only the scheduler. Let me go through the PR really quick and get back to u if we need to stress test this. Overall I feel pretty weird about releasing a version with known performance regression, if it is there. Maybe we haven't yet defined the baselines for Airflow but scheduler performance IMO should be one of them. Does it make sense?
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
