[ 
https://issues.apache.org/jira/browse/HADOOP-18105?focusedWorklogId=774786&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-774786
 ]

ASF GitHub Bot logged work on HADOOP-18105:
-------------------------------------------

                Author: ASF GitHub Bot
            Created on: 25/May/22 19:37
            Start Date: 25/May/22 19:37
    Worklog Time Spent: 10m 
      Work Description: mukund-thakur commented on PR #4263:
URL: https://github.com/apache/hadoop/pull/4263#issuecomment-1137774197

   > deciding how to handle an unknown buffer being returned is the key one
   What do you mean by this? I don't think there is a way to know whether the 
buffer being returned currently through putBuffer() was part of this pool or 
not. 
   Right now the puBuffer() call succeeds if I create a random buffer directly 
in my tests. 
   




Issue Time Tracking
-------------------

    Worklog Id:     (was: 774786)
    Time Spent: 3h 10m  (was: 3h)

> Implement a variant of ElasticByteBufferPool which uses weak references for 
> garbage collection.
> -----------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-18105
>                 URL: https://issues.apache.org/jira/browse/HADOOP-18105
>             Project: Hadoop Common
>          Issue Type: Sub-task
>          Components: common, fs
>            Reporter: Mukund Thakur
>            Assignee: Mukund Thakur
>            Priority: Major
>              Labels: pull-request-available
>          Time Spent: 3h 10m
>  Remaining Estimate: 0h
>
> Currently in hadoop codebase, we have two classes which implements byte 
> buffers pooling.
> One is ElasticByteBufferPool which doesn't use weak references and thus could 
> cause memory leaks in production environment. 
> Other is DirectBufferPool which uses weak references but doesn't support 
> caller's preference for either on-heap or off-heap buffers. 
>  
> The idea is to create an improved version of ElasticByteBufferPool by 
> subclassing it ( as it is marked as public and stable and used widely in hdfs 
> ) with essential functionalities required for effective buffer pooling. This 
> is important for the parent Vectored IO work.



--
This message was sent by Atlassian Jira
(v8.20.7#820007)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to