Henry2SS commented on PR #7476: URL: https://github.com/apache/incubator-gluten/pull/7476#issuecomment-2409877070
> > The performance is 1.x times faster. > > @Henry2SS Do you mind share the concrete performance improvement? My pleasure. There is a benchmark testing CAST VARCHAR to BIGINT. The sql is like `INSERT INTO tmp_table as select CAST(string_col as BIGINT) from table`. And the table has 2,879,966,589 rows. We used gperftool to find that there is a function that takes up the most of the execution time in benchmark. Before removing trim node, the total time of `ProjectExecTransformer` is 17.3 min ,dividing into 515 tasks. After removing trim node, the total time of `ProjectExecTransformer` is 7.5 min ,dividing into 515 tasks. So it is more than 1.x times faster. -- 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]
