[
https://issues.apache.org/jira/browse/HIVE-7526?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chao updated HIVE-7526:
-----------------------
Attachment: HIVE-7526.3.patch
An attempt to fix the last patch by moving groupBy op to ShuffleTran.
Also, since now SparkTran::transform may have input/output value types other
than BytesWritable, we need to make it generic as well..
Also added a CompTran class, which is basically a composition of
transformations. It offers better type compatibility than ChainedTran.
This is NOT the perfect solution, and may subject to further change.
> Research to use groupby transformation to replace Hive existing
> partitionByKey and SparkCollector combination
> -------------------------------------------------------------------------------------------------------------
>
> Key: HIVE-7526
> URL: https://issues.apache.org/jira/browse/HIVE-7526
> Project: Hive
> Issue Type: Task
> Components: Spark
> Reporter: Xuefu Zhang
> Assignee: Chao
> Attachments: HIVE-7526.2.patch, HIVE-7526.3.patch, HIVE-7526.patch
>
>
> Currently SparkClient shuffles data by calling paritionByKey(). This
> transformation outputs <key, value> tuples. However, Hive's ExecMapper
> expects <key, iterator<value>> tuples, and Spark's groupByKey() seems
> outputing this directly. Thus, using groupByKey, we may be able to avoid its
> own key clustering mechanism (in HiveReduceFunction). This research is to
> have a try.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
