When running the mapping with HDFS as source, the storage handler can be used against Hive when a Hive table is used as a source.

The following figure shows the mapping with an Aggregator transformation:

The mapping reads a 750 GB file, applies an expression to each row of data, and stores the aggregated result in a file.

To test performance, the mapping is run with the Informatica storage handler that uses HDFS as source and target. The mapping is also run with the Hive storage handler that uses Hive as source and target.