Dandandan commented on issue #20324: URL: https://github.com/apache/datafusion/issues/20324#issuecomment-4060338616
Seems removing the current "prefetching" helps fixing part of the query 6 performance regression as well (https://github.com/apache/datafusion/pull/20916): ``` │ QQuery 6 │ 11.25 / 13.69 ±3.56 / 20.75 ms │ 7.63 / 8.66 ±0.70 / 9.82 ms │ +1.58x faster │ `` This seems to show that in this case `open` happens too early (so it has to read+decode more than a more lazy approach). This is perhaps something we can solve with the morsel-based planner as well. -- 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]
