To execute batch operations, the driver can use either of the following batch mechanisms:
The Apache Hive native batch mechanism. The native batch mechanism returns individual update counts for each statement or parameter set in the batch as required by the JDBC 3.0 specification.
A mechanism internal to the driver that uses a parameterized multi-row insert statement. The multi-row insert batch mechanism returns only the total number of update counts in the batch, but provides substantial performance gains when performing batch inserts. The multi-row insert batch mechanism only applies to Insert statements for batch executes called with PreparedStatement objects.
The driver's batch mechanism behavior can configured using the BatchMechanism connection property. By default, the driver uses the multi-row insert batch mechanism.