https://community.cloudera.com/t5/Support-Questions/query-hive-tables-with-spark-sql/m-p/145381#M107949 <P>+1 azeltov. We can take that one step further an make the Hive tables available through Spark SQL via JDBC from outside the cluster. </P><P>SPARK_HOME/sbin/start-thriftserver.sh --hiveconf hive.server2.thrift.port={port to listen} --hiveconf hive.server2.thrift.bind.host={host to bind to} --master yarn-client</P><P>This will start a HiveServer2 instance that has access to the meta store but will turn SQL into Spark instruction sets and RDDs under the covers. You should now be able to use a HiveServer compliant JDBC driver to connect and access the power of SparkSQL and yet leverage all of the existing investment and assets in Hive.</P> Wed, 30 Mar 2016 00:42:11 GMT vvaks 2016-03-30T00:42:11Z