[
https://issues.apache.org/jira/browse/HUDI-8468?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17921648#comment-17921648
]
Sagar Sumit commented on HUDI-8468:
-----------------------------------
How can we avoid per record overhead in CUSTOM merge mode given that we need to
load the custom payload class based on reflection and payload API is tightly
coupled to record? Even with delegation pattern, what would
`CustomBasedExpressionPayload` actually do if not load the custom payload?
> Merge into always uses eventtime ordering
> ------------------------------------------
>
> Key: HUDI-8468
> URL: https://issues.apache.org/jira/browse/HUDI-8468
> Project: Apache Hudi
> Issue Type: Sub-task
> Components: spark-sql
> Reporter: Jonathan Vexler
> Assignee: sivabalan narayanan
> Priority: Blocker
> Labels: pull-request-available
> Fix For: 1.0.1
>
> Original Estimate: 8h
> Remaining Estimate: 8h
>
> MIT uses eventtime ordering even when the payload is overwite.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
