HyukjinKwon commented on code in PR #41563:
URL: https://github.com/apache/spark/pull/41563#discussion_r1231605943
##########
python/pyspark/worker.py:
##########
@@ -531,9 +528,10 @@ def func(_, it):
f"the 'terminate' method: {str(e)}"
)
- ser = BatchedSerializer(CPickleSerializer(), 100)
+ # Each row is a group so do not batch but send one by one.
Review Comment:
Yeah it would does because the batch is too small - meaning that I/O happens
more with less buffering. That's why we need a better implementation that
creates a batch based on the actual row number.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
