[ https://issues.apache.org/jira/browse/KUDU-2637?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16719564#comment-16719564 ]
Alexey Serbin commented on KUDU-2637: ------------------------------------- Probably, it makes sense to add some data-backed explanation to illustrate that actual leader-related overhead is not that big. Running {{kudu perf loadgen}} workload against a small Kudu cluster (even with all tablet servers at the same node) and capturing metrics like {{cpu_utime}}, {{cpu_stime}}, {{threads_running}} and a snapshot of the {{MALLOC}} section of the {{/memz}} page generated by embedded Web server might be good enough. > Add a note about leadership imbalance in the faq > ------------------------------------------------ > > Key: KUDU-2637 > URL: https://issues.apache.org/jira/browse/KUDU-2637 > Project: Kudu > Issue Type: Improvement > Components: documentation > Reporter: Grant Henke > Priority: Minor > > There have been a few questions on leadership imbalance and whether or not it > is important to monitor and fix. We should update the FAQ section to address > this. -- This message was sent by Atlassian JIRA (v7.6.3#76005)