[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15383601#comment-15383601
]
ramkrishna.s.vasudevan commented on HBASE-14613:
------------------------------------------------
bq.What about the CMS vs G1GC discussion above?
With pool we were able to tune the CMS and were able to see better performance.
In case of G1GC, because we have offheap now and having a Buffer pool helps
reduce the heap that is required and the tuning of G1GC helps to have a much
predictable throughput. Will try out the same option with onheap and pooling.
Will get back on that here.
> Remove MemStoreChunkPool?
> -------------------------
>
> Key: HBASE-14613
> URL: https://issues.apache.org/jira/browse/HBASE-14613
> Project: HBase
> Issue Type: Sub-task
> Reporter: Lars Hofhansl
> Priority: Minor
> Attachments: 14613-0.98.txt, gc.png, writes.png
>
>
> I just stumbled across MemStoreChunkPool. The idea behind is to reuse chunks
> of allocations rather than letting the GC handle this.
> Now, it's off by default, and it seems to me to be of dubious value. I'd
> recommend just removing it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
