pengbo commented on issue #24349: [Spark-27416][SQL]UnsafeMapData & UnsafeArrayData Kryo serialization breaks when two machines have different Oops size URL: https://github.com/apache/spark/pull/24349#issuecomment-482184576 > It would be for performance, for the same reason all the basic spark classes are registered; otherwise Kryo serializes the class name with each instance. Why wouldn't it matter? maybe I'm missing something The kyro & java serialization are using the same method that's why I think it would makes no different. I am not quite familiar with how kyro is using in spark yet, let me know what i'm missing and the following step I can do. Thanks
---------------------------------------------------------------- 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 --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
