bshashikant commented on issue #578: HDDS-3053. Decrease the number of the chunk writer threads URL: https://github.com/apache/hadoop-ozone/pull/578#issuecomment-597591486 > > did u check the pending request queue in the leader? > > No I didn't. Why is it interesting? > > > how many mappers were used for the test? > > 92 (see the link for this and all the other parameters. With lesser no of threads, if the no pending tasks keep on building on the chunk executor queue bcoz of lesser no of threads available for work, once it hits the max limit, the main thread itself starts executing the write call which degrades the performance as per the caller run policy set in the writeChunkExecutor as observed in our tests. @elek , if you really think this seems to be causing issues in tests , we can change it to a lower value.
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: ozone-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: ozone-issues-h...@hadoop.apache.org
