[ https://issues.apache.org/jira/browse/KUDU-1587?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17186284#comment-17186284 ]
ASF subversion and git services commented on KUDU-1587: ------------------------------------------------------- Commit c6d438ab417009e8007a1de274178d0bcf0dfb63 in kudu's branch refs/heads/master from Alexey Serbin [ https://gitbox.apache.org/repos/asf?p=kudu.git;h=c6d438a ] [tserver] add test to reproduce KUDU-1587 conditions Added a test to reproduce conditions described in KUDU-1587. As of now, the test is disabled: it will be enabled once KUDU-1587 is addressed. Change-Id: I515a1b26152680ee9b9361afcf84fec39b8f962d Reviewed-on: http://gerrit.cloudera.org:8080/16312 Tested-by: Kudu Jenkins Reviewed-by: Andrew Wong <aw...@cloudera.com> > Memory-based backpressure is insufficient on seek-bound workloads > ----------------------------------------------------------------- > > Key: KUDU-1587 > URL: https://issues.apache.org/jira/browse/KUDU-1587 > Project: Kudu > Issue Type: Bug > Components: tserver > Affects Versions: 0.10.0 > Reporter: Todd Lipcon > Assignee: Alexey Serbin > Priority: Critical > Labels: roadmap-candidate > Attachments: graph.png, queue-time.png > > > I pushed a uniform random insert workload from a bunch of clients to the > point that the vast majority of bloom filters no longer fit in buffer cache, > and the compaction had fallen way behind. Thus, every inserted row turns into > 40+ seeks (due to non-compact data) and takes 400-500ms. In this kind of > workload, the current backpressure (based on memory usage) is insufficient to > prevent ridiculously long queues. -- This message was sent by Atlassian Jira (v8.3.4#803005)