jerqi commented on issue #736: URL: https://github.com/apache/incubator-uniffle/issues/736#issuecomment-1473388812
> > Introduce a new configuration rss.server.hdfs.speed. We record the timestamp which the partition start to receive data. And then we record the timestamp which the partition start to flush data. We can use two timestamps to calcuate partition receive data speed. We use (partition receive data speed / HDFS write speed) to calculate how many threads which we need to write HDFS data. We use the number to create HDFS writer. > > It looks too complex, the initial goal is to solve the single partition lock problem. Does above solution will solve this? Yes, we don't need to adjust the parameter by hand. -- 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: dev-unsubscr...@uniffle.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
