General Restrictions for Hive Targets
You can use the Update Strategy transformation on the Hadoop distributions that support Hive ACID.
You can also use the Update Strategy transformation to write mapping results to JDBC-compliant relational targets.
The Update Strategy transformation does not forward rejected rows to the next transformation when the target is a Hive table or JDBC-compliant table.
If the Update Strategy transformation receives multiple update rows for the same primary key value, the transformation selects one random row to update the target.
If multiple Update Strategy transformations write to different instances of the same target, the target data might be unpredictable.
If the mapping runs on the Spark engine, you can choose the Use Hive Merge option. The option has the following restrictions:
A single row for delete or update cannot match multiple rows in the target. When the mapping violates this restriction, the mapping fails with a runtime error.
If you configure the Update Strategy expression to update partitioning or bucketing columns, the mapping ignores the Hive MERGE option and does not update the columns.
The Developer tool and the Data Integration Service do not validate against these restrictions. If the expression or the mapping violates these restrictions, the mapping might run, but the results will not be as expected.
Hive targets always perform Update as Update operations. Hive targets do not support Update Else Insert or Update as Insert.