> in favour of Hive, Spark Well, Hive and Spark create MapReduce jobs to satisfy your queries: it's still in the background, but you don't have to think about

> in favour of Hive, Spark Well, Hive and Spark create MapReduce jobs to satisfy your queries: it's still in the background, but you don't have to think about it (as much).