[jira] [Commented] (TEZ-2879) While grouping splits, allow an alternate list of preferred locations to be provided per split

["Gopal V (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/TEZ-2879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14992889#comment-14992889 ] 

Gopal V commented on TEZ-2879:
------------------------------

[~sseth]: the API LGTM - +1.

This mechanism for location pinning can also benefit filesystems like WASB on Azure.

> While grouping splits, allow an alternate list of preferred locations to be provided per split
> ----------------------------------------------------------------------------------------------
>
>                 Key: TEZ-2879
>                 URL: https://issues.apache.org/jira/browse/TEZ-2879
>             Project: Apache Tez
>          Issue Type: Improvement
>            Reporter: Siddharth Seth
>            Assignee: Siddharth Seth
>         Attachments: TEZ-2879.1.txt, TEZ-2879.1.wip.txt
>
>
> Split locations - at least for FileInputSplits - are generally tied to the location on HDFS where the split resides.
> There are situations in which this location is not necessarily the best location to process this split.
> e.g.
> Clusters where compute and storage are separate.
> Systems which cache data - cache affinity is more important.
> Providing an alternate list of preferred locations allows grouping to the preferred locations, instead of always grouping based on the locations specified in the split.



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