You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@giraph.apache.org by "Hudson (JIRA)" <ji...@apache.org> on 2016/04/05 01:45:25 UTC

[jira] [Commented] (GIRAPH-1046) Add a way to synchronize full GC calls across workers

    [ https://issues.apache.org/jira/browse/GIRAPH-1046?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15225297#comment-15225297 ] 

Hudson commented on GIRAPH-1046:
--------------------------------

FAILURE: Integrated in Giraph-trunk-Commit #1594 (See [https://builds.apache.org/job/Giraph-trunk-Commit/1594/])
GIRAPH-1046: Add a way to synchronize full GC calls across workers (majakabiljo: [http://git-wip-us.apache.org/repos/asf?p=giraph.git&a=commit&h=24bed1a9bdbe3a1edd46bc3b2af83ee385e041a9])
* giraph-core/src/main/java/org/apache/giraph/bsp/BspService.java
* src/site/xdoc/options.xml
* giraph-core/src/main/java/org/apache/giraph/worker/MemoryObserver.java
* giraph-core/src/main/java/org/apache/giraph/worker/BspServiceWorker.java


> Add a way to synchronize full GC calls across workers
> -----------------------------------------------------
>
>                 Key: GIRAPH-1046
>                 URL: https://issues.apache.org/jira/browse/GIRAPH-1046
>             Project: Giraph
>          Issue Type: New Feature
>            Reporter: Maja Kabiljo
>            Assignee: Maja Kabiljo
>
> In applications which use memory more heavily, we can see full GC pauses happening on different workers at different times, and each of these is causing some delay because other workers are often waiting on something from the worker in GC (closing open requests, finishing superstep, etc). Having a way to coordinate when full GCs are called could help them have less effect on job performance.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)