924060929 commented on issue #7212: URL: https://github.com/apache/incubator-doris/issues/7212#issuecomment-982365991
> 这个查询规划如果顶层 join 改为shuffle 后，不是会传输更多的数据？为何从 broadcast 改为shuffle 后反而效率变高了？ 新版runtime filter默认type是in，在shuffle join中in会隐式转成bloom filter，提高了性能。但broadcast join中runtime filter还是保持了in，所以相比性能较低。 后续尝试增加一个in_or_bloomfilter的类型，让shuffle/broadcast join的runtime filter行为保持一致，并且去掉在in类型下隐式转换bloom filter的行为 -- 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]
