You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@falcon.apache.org by "Ajay Yadava (JIRA)" <ji...@apache.org> on 2016/01/27 21:38:40 UTC

[jira] [Commented] (FALCON-1785) NFS as feed entity location

    [ https://issues.apache.org/jira/browse/FALCON-1785?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15120124#comment-15120124 ] 

Ajay Yadava commented on FALCON-1785:
-------------------------------------

Thank you for reporting this [~aahn]. Most of this makes sense to me. 

Can you please elaborate on point 1? If possible, can you share what schedulers you use as it will help to see what kind of integration is suitable? 



> NFS as feed entity location 
> ----------------------------
>
>                 Key: FALCON-1785
>                 URL: https://issues.apache.org/jira/browse/FALCON-1785
>             Project: Falcon
>          Issue Type: New Feature
>            Reporter: Andrew Ahn
>              Labels: bank
>   Original Estimate: 672h
>  Remaining Estimate: 672h
>
> Falcon - integration with extenal NFS (as source). This will allow Falcon to create process to use NFS location for source of data movement.   As a consequence:
> 1   Flag based integration with external enterprise schedulers can be acheived by checking outside of Hadoop - specifically a NFS share.
> 2   Ability to use NFS location as the input to a process
> 3   Ability to use NFS location as the output to a process
> 4   As a stretch - Falcon / Oozie should be able to evict data on a NFS location.



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