[
https://issues.apache.org/jira/browse/HBASE-14098?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14631708#comment-14631708
]
Elliott Clark commented on HBASE-14098:
---------------------------------------
{code}
vm.dirty_background_ratio = 3
vm.dirty_ratio = 20
{code}
The interesting part is that these stalls aren't really caused by dirty pages.
There's only 400mb of dirty pages. These stalls seem to be caused by thrashing
the cache. Too much reading/writing actually causes reads to happen on the root
disk to re-page in executables.
> Allow dropping caches behind compactions
> ----------------------------------------
>
> Key: HBASE-14098
> URL: https://issues.apache.org/jira/browse/HBASE-14098
> Project: HBase
> Issue Type: Bug
> Components: Compaction, hadoop2, HFile
> Affects Versions: 2.0.0, 1.3.0
> Reporter: Elliott Clark
> Assignee: Elliott Clark
> Fix For: 2.0.0, 1.3.0
>
> Attachments: HBASE-14098-v1.patch, HBASE-14098.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
