[
https://issues.apache.org/jira/browse/HBASE-21921?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Hu updated HBASE-21921:
-----------------------------
Resolution: Fixed
Hadoop Flags: Reviewed
Status: Resolved (was: Patch Available)
> Notify users if the ByteBufAllocator is always allocating ByteBuffers from
> heap which means the increacing GC pressure
> ----------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-21921
> URL: https://issues.apache.org/jira/browse/HBASE-21921
> Project: HBase
> Issue Type: Sub-task
> Reporter: Zheng Hu
> Assignee: Zheng Hu
> Priority: Minor
> Attachments: HBASE-21921.HBASE-21879.v01.patch,
> HBASE-21921.HBASE-21879.v02.patch, HBASE-21921.HBASE-21879.v03.patch,
> jmx-metrics.png, web-ui.png
>
>
> As the javadoc of ByteBuffAllocator says:
> {code}
> There's possible that the desired memory size is large than ByteBufferPool
> has, we'll downgrade to allocate ByteBuffers from heap which meaning the GC
> pressure may increase again. Of course, an better way is increasing the
> ByteBufferPool size if we detected this case.
> {code}
> So I think we need some messages to remind the user that an larger
> ByteBufferPool size may be better if the allocator allocate ByteBuffer from
> heap frequently.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
