Open jackye1995 opened 1 year ago
+1 for this, it will be awesome if we could make the retry pluggable as well :) !
This issue has been automatically marked as stale because it has been open for 180 days with no activity. It will be closed in next 14 days if no further activity occurs. To permanently prevent this issue from being considered stale, add the label 'not-stale', but commenting on the issue is preferred when possible.
This issue has been closed because it has not received any activity in the last 14 days since being marked as 'stale'
Feature Request / Improvement
Currently we just use exponential backoff. In practice when there are many concurrent writes, the conflict rate is still high even with jitter. There are some other algorithms like AIMD that was proven to be effective, and might be worth considering to add to Tasks to retry commits.
Reference: https://docs.aws.amazon.com/emr/latest/ReleaseGuide/emr-spark-emrfs-retry.html
Query engine
None