[
https://issues.apache.org/jira/browse/TEZ-1610?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14240315#comment-14240315
]
Hitesh Shah edited comment on TEZ-1610 at 12/9/14 11:47 PM:
------------------------------------------------------------
[~rajesh.balamohan] Regarding SHUFFLE_TIME_AS_PERCENTAGE ( and to an extent all
the other new counters), how are these counters meant to be used by users when
viewing counters at the DAG level ( which will be the common scenario ) and
maybe at the vertex level?
Also, for a user using partitioned ordered inputs and outputs, what is the user
meant to understand by "shuffle" i.e shuffled data, shuffle side, etc. ?
Likewise, what is a "fetcher" for a user?
was (Author: hitesh):
[~rajesh.balamohan] Regarding SHUFFLE_TIME_AS_PERCENTAGE ( and to an extent all
the other new counters), how are these counters meant to be used by users when
viewing counters at the DAG level ( which will be the common scenario ) and
maybe at the vertex level?
Also, for a user using partitioned ordered inputs and outputs, what is the user
meant to understand by "shuffle" i.e shuffled data, shuffle side, etc. ?
> additional task counters for fetchers
> -------------------------------------
>
> Key: TEZ-1610
> URL: https://issues.apache.org/jira/browse/TEZ-1610
> Project: Apache Tez
> Issue Type: Bug
> Reporter: Rajesh Balamohan
> Assignee: Rajesh Balamohan
> Attachments: TEZ-1610.1.patch, TEZ-1610.2.patch
>
>
> - ShuffleFinishTime (per source)
> - Merge time (depending on broadcast/scatter-gather shuffle)
> This would be helpful in determining when shuffle started/ended for different
> sources in a task.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
