hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ramkrishna.s.vasudevan (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (HBASE-14613) Remove MemStoreChunkPool?
Date Thu, 24 Mar 2016 05:04:25 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15208316#comment-15208316
] 

ramkrishna.s.vasudevan edited comment on HBASE-14613 at 3/24/16 5:03 AM:
-------------------------------------------------------------------------

bqThat means that if you're promoting a decent amount of objects but not getting enough mixed
gen collections, it's possible to get pauses even though there is a lot of memory left.
[~eclark]
I repeated the experiment and I can see that the case where we have No MSLAB and no chunk
pool the GC pause is infact more and the case where we have only MSLAB and no chunk pool the
GC pause is lesser but the through put is higher in case the NO MSLAB case. What could be
the reason for that?  Does your results also go with this justification? 


was (Author: ram_krish):
bqThat means that if you're promoting a decent amount of objects but not getting enough mixed
gen collections, it's possible to get pauses even though there is a lot of memory left.
[~eclark]
I repeated the experiment and I can see that the case where we have No MSLAB and no chunk
pool the GC pause is infact more and the case where we have only MSLAB and no chunk pool the
GC pause is lesser but the through put is higher than the NO MSLAB case. What could be the
reason for that?  Does your results also go with this justification? 

> 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