Kwafoor commented on pull request #18498: URL: https://github.com/apache/flink/pull/18498#issuecomment-1064683134
> I don't see what that gives you though. if the existing load metric is close to 1 you already know that an increase of the number of CPU cores could make sense. Multiplying the load by a constant factor doesn't really get you anything imo. If you _actually_ want to determine CPU bottlenecks, then https://nightlies.apache.org/flink/flink-docs-release-1.14/docs/ops/metrics/#system-resources is a better suited anyway. If multiple applications share resources in cluster like yarn, overall CPU usage on the machine and cpu load is meaningless.All we need to know is how much resources flink used and how efficiency it is. On one machine is a constant factor but in cluster machine core is different. -- 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: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org