what are the key differences between hive QL and Spark-SQL ? which one is better for performance and why ?
Hive QL is a query language, Spark SQL too is a query language.
The queries written in Hive QL are executed as a Map Reduce job, which uses the disk(SSD/Magnetic disks) to store intermediate results which therefore takes more time to execute.
The queries written in Spark SQL are executed as Spark jobs that use memory(RAM) to store and perform operations on input data. Therefore the performance is better than HIVE queries.