KevinyhZou commented on issue #12459: URL: https://github.com/apache/gluten/issues/12459#issuecomment-4924641525
This seems different from our current implementation: the input `RowData` read from flink source or other operators, and then convert to velox data structure `ChangelogRowVector`, however, current velox has implemented some native operators, which can directly read and handle data. so, should we consider about offloading the whole pipeline to velox, like: reading input in velox sideļ¼and then convert it to `ChangelogRowVector` directly? @ggjh-159 -- 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]
