[ https://issues.apache.org/jira/browse/HIVE-7333?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14209297#comment-14209297 ]
Rui Li commented on HIVE-7333: ------------------------------ Hi [~klonikar], Thanks for your interest. When I did the tests, I only checked whether spark can support the various hive file formats. I didn't dive into the code details either. To answer your question, my guess is that both hive table and spark RDD are nothing but a bunch of hadoop files at the very bottom. How the file is read is the job of specific RecordReader/SerDe of that file format. At the higher level both hive table and RDD are row-wise. [~xuefuz] please correct me if I misunderstand. Although I didn't find any issue during this JIRA, please feel free to report any problems you found. Thanks! > Create RDD translator, translating Hive Tables into Spark RDDs [Spark Branch] > ----------------------------------------------------------------------------- > > Key: HIVE-7333 > URL: https://issues.apache.org/jira/browse/HIVE-7333 > Project: Hive > Issue Type: Sub-task > Components: Spark > Reporter: Xuefu Zhang > Assignee: Rui Li > Labels: Spark-M1 > > Please refer to the design specification. -- This message was sent by Atlassian JIRA (v6.3.4#6332)