yugan95 commented on PR #56542:
URL: https://github.com/apache/spark/pull/56542#issuecomment-4727139308

   @HeartSaVioR Thanks for the follow-up questions.
   
   To clarify the shard infrastructure: it is fully inlined in the executor 
process — just an additional Netty server and a few RPC endpoints started 
inside `SparkEnv`. There is no separate process, no external service, and no 
additional deployment step. When `spark.shard.enabled=false` (the default), 
none of this code is initialized and there is zero overhead.
   
   The reason it defaults to off is that the Netty server opens an additional 
port on each executor. Keeping it opt-in felt more appropriate rather than 
starting a new listener by default.
   
   I understand the concern about scope and I'm open to drafting a SPIP. Will 
wait to hear from the SQL area maintainers as well.
   


-- 
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]

Reply via email to