wForget commented on PR #12067: URL: https://github.com/apache/gluten/pull/12067#issuecomment-4621907449
> Hi @jaylisde I understand the checksum computation was introduced to resolve the data consistency issue with task retries, and we do lack this support for spark 4.0 in Gluten. But since Gluten's hash-based shuffle handles the data in columnar format, I wonder if it's possible to implement this based on each column buffer? The column buffer of shuffle write data is unstable; when upstream recalculates, using it to calculate checksun cannot guarantee that the same value will be generated. -- 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]
