[
https://issues.apache.org/jira/browse/HIVE-23375?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17100696#comment-17100696
]
Panagiotis Garefalakis commented on HIVE-23375:
-----------------------------------------------
Hey [~kgyrtkirk] thanks for the comments!
Adding a memory estimation counter would make sense indeed, can work on that as
a follow-up.
Currently, these counters are aggregated in vertex level so yes it would be the
cumulative load_time which gives you some information but not the whole picture
I agree.
Going task level would be too much so let me dig a bit deeper on Tez level to
check how we can introduce a counter that tracks a distribution of values.
> Track MJ HashTable Load time
> ----------------------------
>
> Key: HIVE-23375
> URL: https://issues.apache.org/jira/browse/HIVE-23375
> Project: Hive
> Issue Type: Improvement
> Reporter: Panagiotis Garefalakis
> Assignee: Panagiotis Garefalakis
> Priority: Minor
> Labels: pull-request-available
> Attachments: HIVE-23375.01.patch
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Introduce TezCounter to track MJ HashTable Load time
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
