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

Posted to dev@hama.apache.org by "Suraj Menon (JIRA)" <ji...@apache.org> on 2014/02/13 11:14:19 UTC

[jira] [Updated] (HAMA-636) Confined recovery

     [ https://issues.apache.org/jira/browse/HAMA-636?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Suraj Menon updated HAMA-636:
-----------------------------

    Labels: gsoc2014 java  (was: )

> Confined recovery
> -----------------
>
>                 Key: HAMA-636
>                 URL: https://issues.apache.org/jira/browse/HAMA-636
>             Project: Hama
>          Issue Type: Sub-task
>          Components: bsp core, messaging
>            Reporter: Edward J. Yoon
>              Labels: gsoc2014, java
>
> "Confined recovery" mentioned in Pregel paper can be used to improve the cost and latency of recovery. 
> In addition to the existing HDFS checkpoints,1) the tasks log outgoing messages to local filesystem for each superstep (See disk queue). When a task fails, 2) it reverts to the last checkpoint. 3) Other tasks re-send messages sent to failed task at each superstep occurring after the last checkpoint.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)