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

Posted to dev@storm.apache.org by "Rick Kellogg (JIRA)" <ji...@apache.org> on 2015/10/05 03:26:27 UTC

[jira] [Updated] (STORM-882) Dist Cache: Have HDFS blob store download starting at a random block

     [ https://issues.apache.org/jira/browse/STORM-882?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Rick Kellogg updated STORM-882:
-------------------------------
    Component/s: storm-core

> Dist Cache: Have HDFS blob store download starting at a random block
> --------------------------------------------------------------------
>
>                 Key: STORM-882
>                 URL: https://issues.apache.org/jira/browse/STORM-882
>             Project: Apache Storm
>          Issue Type: Improvement
>          Components: storm-core
>            Reporter: Robert Joseph Evans
>
> To reduce the possibility of overloading a datanode when downloading large dist cache items we should start downloading the blob at a random block, and then loop back to the beginning of the file to finish.  STORM-881 provides randomizing the order of the files downloaded, but on an update of a large blob this may become important.



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