[
https://issues.apache.org/jira/browse/HIVE-6222?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Remus Rusanu updated HIVE-6222:
-------------------------------
Status: Patch Available (was: Open)
.3.Patch addresses the test failures. Incorrect comparison in
checkHashEfficiency was triggering switch to streaming mode on first row
processed. While the fix addresses the problem, the results diff also showed
that there are rounding diffs between streamign mode (agg done using
UnsignedInt128) vs. streaming mode (agg done on reduce side, using
HiveDecimal). This is similar to the issues HIVE-6511 exposed and I'll open a
separate JIRA to address it.
> Make Vector Group By operator abandon grouping if too many distinct keys
> ------------------------------------------------------------------------
>
> Key: HIVE-6222
> URL: https://issues.apache.org/jira/browse/HIVE-6222
> Project: Hive
> Issue Type: Sub-task
> Components: Query Processor
> Affects Versions: 0.13.0
> Reporter: Remus Rusanu
> Assignee: Remus Rusanu
> Priority: Minor
> Labels: vectorization
> Attachments: HIVE-6222.1.patch, HIVE-6222.2.patch, HIVE-6222.3.patch
>
>
> Row mode GBY is becoming a pass-through if not enough aggregation occurs on
> the map side, relying on the shuffle+reduce side to do the work. Have VGBY do
> the same.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
