https://community.cloudera.com/t5/Support-Questions/Some-nodes-are-way-slower-on-HDFS-scan-then-the-other-ones/m-p/307942#M223418 <P>"... or because memory is running low or because the extra scanner threads are not needed to keep up with the consumer from the scan node." is how I should've finished that.<BR /><BR />Not sure what happened there, I did hit the main points but truncated the last sentence.<BR /><BR />That part of Impala has had a lot of improvements for performance and observability (i.e. more info in the profile) since CDH5.10, FWIW, I'd guess on a later version this wouldn't be a problem or would be easier to debug at least.</P> Thu, 17 Dec 2020 21:09:45 GMT Tim Armstrong 2020-12-17T21:09:45Z