[
https://issues.apache.org/jira/browse/TEZ-624?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13876754#comment-13876754
]
Siddharth Seth commented on TEZ-624:
------------------------------------
Should getTaskIndex / getOutputIndex be setup as explicit APIs. If these are
exposed, we should likely get rid of getUniqueIdentifier, and let users
construct that on their own (and expose InputIndex as well for consistency).
Alternately, uniqueIdentifier could just be changed to be
vertexId_IOIndex_taskIndex.
The Javadoc on those methods is a little confusing - getOutputIndex - is that
for the specific Vertex, or for the entire DAG. Also, are there any semantics
as to what these IDs will be with relation to how a user setup the DAG - or are
they just some kind of unique identifiers ?
NUMBER_FORMAT likely needs to be thread specific, especially if there's
multiple Outputs. Separate NUMBER_FORMAT for vertexId - 5 is rather long?
> Fix MROutput to support multiple outputs to the same output location
> --------------------------------------------------------------------
>
> Key: TEZ-624
> URL: https://issues.apache.org/jira/browse/TEZ-624
> Project: Apache Tez
> Issue Type: Bug
> Reporter: Hitesh Shah
> Assignee: Bikas Saha
> Attachments: TEZ-624.1.patch, TEZ-624.2.patch, TEZ-624.3.patch
>
>
> For users using multiple MROutput instances on a single vertex
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
