hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ryan Brush (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-6322) Unnecessary creation of finalizers in HTablePool
Date Wed, 04 Jul 2012 21:14:34 GMT

    [ https://issues.apache.org/jira/browse/HBASE-6322?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13406757#comment-13406757
] 

Ryan Brush commented on HBASE-6322:
-----------------------------------

My apologies...I had only run the tests around HTablePool since I had run into some apparently
unrelated test failures in a full build.  (And I didn't expect it to be included in the build
so quickly. ;)  It looks like we'll need to do some refactoring in REST server's RegionResource
for this to apply cleanly, specifically the call to HTable.getRegionsInfo which requires the
downcast (and is deprecated, anyway).

I'm not deeply familiar with this part of the code base and probably won't be able to dig
into it today, but can get back to it in the next couple days, as well as making sure there
aren't any further regressions caused by this change.
                
> Unnecessary creation of finalizers in HTablePool
> ------------------------------------------------
>
>                 Key: HBASE-6322
>                 URL: https://issues.apache.org/jira/browse/HBASE-6322
>             Project: HBase
>          Issue Type: Bug
>          Components: client
>    Affects Versions: 0.92.0, 0.92.1, 0.94.0
>            Reporter: Ryan Brush
>             Fix For: 0.92.2
>
>         Attachments: HBASE-6322-0.92.1.patch, HBASE-6322-trunk.1.patch
>
>
> From a mailing list question:
> While generating some load against a library that makes extensive use of HTablePool in
0.92, I noticed that the largest heap consumer was java.lang.ref.Finalizer.  Digging in, I
discovered that HTablePool's internal PooledHTable extends HTable, which instantiates a ThreadPoolExecutor
and supporting objects every time a pooled HTable is retrieved.  Since ThreadPoolExecutor
has a finalizer, it and its dependencies can't get garbage collected until the finalizer runs.
 The result is by using HTablePool, we're creating a ton of objects to be finalized that are
stuck on the heap longer than they should be, creating our largest source of pressure on the
garbage collector.  It looks like this will also be a problem in 0.94 and trunk.
> The easy fix is just to have PooledHTable implement HTableInterface (rather than subclass
HTable), but this does break a unit test that explicitly checks that PooledHTable implements
HTable -- I can only assume this test is there for some historical passivity reason.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message