[
https://issues.apache.org/jira/browse/KUDU-1156?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated KUDU-1156:
------------------------------
Fix Version/s: 0.5.0
> Compactions always preferred above flushes when MRS usage <64MB
> ---------------------------------------------------------------
>
> Key: KUDU-1156
> URL: https://issues.apache.org/jira/browse/KUDU-1156
> Project: Kudu
> Issue Type: Bug
> Components: tserver
> Affects Versions: Private Beta
> Reporter: Todd Lipcon
> Assignee: Jean-Daniel Cryans
> Priority: Critical
> Fix For: 0.5.0
>
> Attachments: Screen Shot 2015-09-18 at 4.44.16 PM.png, graph.png,
> ycsb-load-5nodes_workload-OVERALL-runtime_ms-jobs-runtime.png
>
>
> Looked a bit at the tablet servers in the YCSB cluster today, after the
> cluster had hit about 1TB of data each (about 789 tablets per server). YCSB
> is getting slower and slower, apparently because a lot of RAM is being used
> up by tables that were written days ago. Looking at the maintenance manager
> dashboard, we are prioritizing compactions, and never flushing MRS even from
> tablets that haven't taken an insert in days.
> I think once the total number of tablets grows a bit more, we'll be in a
> situation where we won't flush them, but we're basically out of RAM, and
> everything will grind to a halt.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
