[
https://issues.apache.org/jira/browse/TEZ-3363?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17013346#comment-17013346
]
Syed Shameerur Rahman commented on TEZ-3363:
--------------------------------------------
[~kshukla] Any update on this?
Going through the initial patch, I think using List<NodeId> nodesList will lead
to redundant http calls. Eg: No: of tasks in a vertex is 1000 which were
launched across 5 nodes. Instead of sending 1000 HTTP requests we can make 5
HTTP requests and do some regex matching to delete the shuffle data. Since we
are already matching the regex it will be better to use Set<NodeId> nodesList.
I can pick this if you are not working on this.
cc [~jeagles]
> Delete intermediate data at the vertex level for Shuffle Handler
> ----------------------------------------------------------------
>
> Key: TEZ-3363
> URL: https://issues.apache.org/jira/browse/TEZ-3363
> Project: Apache Tez
> Issue Type: Sub-task
> Reporter: Jonathan Turner Eagles
> Assignee: Kuhu Shukla
> Priority: Major
> Attachments: TEZ-3363.001.patch, TEZ-3363.002.patch
>
>
> For applications like pig where processing times can be very long,
> applications may choose to delete intermediate data for a sub dag. For
> example if a DAG has synced data to HDFS, all upstream intermediate data can
> be safely deleted.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
