1.
SPARK CONF中添加hive-site.xml
hive.metastore.uris
thrift://master:9083
2.
启动hive元数据
hive --metastore >meta.log 2>&1 &
3.
scala>val hiveContext = new org.apache.spark.sql.hive.HiveContext(sc) //hiveContext
scala>hiveContext.sql("use hive") datasource(hive)
scala>hiveContext.sql("show tables").collect.foreach(println)
scala>hiveContext.sql(select count(*) from aaa)
scala>sqlContext.read.json(/aaa/bbb/ccc/ddd.json)
scala>sqlContext.show()// //create dataframe from dataSource
scala>sqlContext.printScheme()Spark SQL on HIVE
原文：http://www.cnblogs.com/wjsshide/p/5560538.html
