jinchengchenghh commented on PR #8273: URL: https://github.com/apache/incubator-gluten/pull/8273#issuecomment-2556107286
Your test key is String before, it use Velox Column to row and row to Velo column, and I have optimized to Column to Arrow and Arrow to column, the performance for string should be improved. And this case may go to regression, suppose udf(a) returns string, hash(udf(a)) + hash(b), change to execute udf(a) in Spark, this will introduce extra arrow to column of udf(a) string result type as _partialProjectAlias column, if totally fallback, the row to column type is int. -- 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]
