Hive on Spark

Hive on Spark is a project to bring together the Hive and Spark communities, and provide a better performing and more robust version of Hive.

Cloudera, Intel, MapR, Databricks, and IBM are working together to bring the project to users. In this talk we'll discuss the motivations, technical overview, implementation challenges, and a demo.

About the Speaker

Szehon Ho is a software engineer in Cloudera and a Hive committer, based in Palo Alto, CA. Prior to this, he was a principal software engineer in Informatica, gaining experience in enterprise software development and the field of data integration. He holds a BS in EECS from UC Berkeley.