[
https://issues.apache.org/jira/browse/HIVE-16369?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Matt McCline updated HIVE-16369:
--------------------------------
Attachment: HIVE-16369.01.patch
Work-in-progress patch.
Spark equivalent of ReduceRecordSource needs conversion to
setNextVectorBatchGroupStatus.
VectorPTFOperator can only handle partition/frame size of 1 batch (1024 rows).
Need to add memory buffering of VRBs and overflow storage on disk.
> Vectorization: Support PTF (Part 1: No Custom Window Framing -- Default Only)
> -----------------------------------------------------------------------------
>
> Key: HIVE-16369
> URL: https://issues.apache.org/jira/browse/HIVE-16369
> Project: Hive
> Issue Type: Bug
> Components: Hive
> Reporter: Matt McCline
> Assignee: Matt McCline
> Priority: Critical
> Attachments: HIVE-16369.01.patch
>
>
> Support vectorization of PTF that doesn't include custom PRECEDING /
> FOLLOWING window frame clauses.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
