Xuefu Zhang created HIVE-8275:
---------------------------------
Summary: Introduce MapInput encapsulating a Hadoop RDD [Spark
Branch]
Key: HIVE-8275
URL: https://issues.apache.org/jira/browse/HIVE-8275
Project: Hive
Issue Type: Sub-task
Components: Spark
Reporter: Xuefu Zhang
Currently the input to a MapWork is an Hadoop RDD. With HIVE-8118, this input
may be shared by multiple MapWorks. Thus, having an object representation of a
map input in SparkPlan will be helpful visually and accurately capture the
meaning of "sharing".
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
