hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Elliott Clark (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-14613) Remove MemStoreChunkPool?
Date Wed, 23 Mar 2016 16:18:25 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15208690#comment-15208690
] 

Elliott Clark commented on HBASE-14613:
---------------------------------------

bq.What could be the reason for that?
We did a lot of investigation to figure out why we needed the off heap block cache even. After
that I didn't do a whole lot of investigation. I just got the numbers and took what worked
better. However I am pretty sure that it's something to do with the fact that putting data
in the older regions necessitates more mixed collections which are much more expensive.

> Remove MemStoreChunkPool?
> -------------------------
>
>                 Key: HBASE-14613
>                 URL: https://issues.apache.org/jira/browse/HBASE-14613
>             Project: HBase
>          Issue Type: Brainstorming
>            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)

Mime
View raw message