We made a comparison of the number of records between Hive on MapReduce and Hive on Spark.And they are in good agreement. But how to ensure that the record values of Hive on MapReduce and Hive on Spark are completely consistent? Do you have any suggestions?
Best wishes. Thanks.
