Bigdata SQL: Upcoming Features in Hive

Still missing from Hive are features such as referential integrity, unique constraints, secondary indexes, and other multiple types of indexes. Hive has had indexes for quite a while now, but their usage in real-life applications has been very minimal. There is no plan to support any of these features in the near future.
The latest version of Hive 2.0 has added the following capabilities from an SQL and performance perspective:
• Metastore support on HBase
• LLAP support
• Support for Hive Procedural Language, which implements procedural SQL for Hive
• Multiple features to support integration of Hive with Apache Spark
The following are some of the major still remaining functionalities to be supported
by Hive:
• Support for Theta join : With this, tables can be joined on nonequality-based conditions on the columns.
• Support for GeoSpatial analytics
• Support for updatable views : Views have existed in Hive for a while, and all relational databases support updatable views with their own rules. Updatable views in Hive can help in some ETL related tasks, especially when the underlying data changes.