[jira] [Comment Edited] (FLUME-2197) Memory Channel has GC issues

["Roshan Naik (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLUME-2197?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14031389#comment-14031389 ] 

Roshan Naik edited comment on FLUME-2197 at 6/14/14 1:38 AM:
-------------------------------------------------------------

During the first 6 mins or so... the total size of allocated objects is still below the max heap size. Once it reaches the max heap size,  the allocation graph flattens. You can see the Green portion does bump up any more. In this state Java is basically doing a minimal GC (tiny peaks) and not a deep gc. So things stay kind of flat.

In the flat section after 6mins, you can see more of the tiny peaks  (small gc cycles) per unit time in the mem ch graph.


was (Author: roshan_naik):
During the first 6 mins or so... the total size of allocated objects is still well below the max heap size. Once it reaches the max heap size,  the allocation graph flattens. You can see the Green portion does bump up any more. In this state Java is basically doing a minimal GC (tiny peaks) and not a deep gc. So things stay kind of flat.

In the flat section after 6mins, you can see more of the tiny peaks  (small gc cycles) per unit time in the mem ch graph.

> Memory Channel has GC issues
> ----------------------------
>
>                 Key: FLUME-2197
>                 URL: https://issues.apache.org/jira/browse/FLUME-2197
>             Project: Flume
>          Issue Type: Bug
>            Reporter: Hari Shreedharan
>            Assignee: Roshan Naik
>         Attachments: mem ch - mem alloc.png, spill ch - mem alloc.png
>
>
> Due to the fact that we use a LinkedBlockingDeque as the backing queue for the MemoryChannel, we end up hitting GC issues more often than we should.



--
This message was sent by Atlassian JIRA
(v6.2#6252)