How do I conditionally distribute data to multiple targets?
You can conditionally distribute data to multiple targets when mapped column data is distributed only for the subset of records that satisfy a user-specified condition. Proper use of this process occurs in cases in which you want to distribute only source records that have a particular date column with a value later than January 1, 2003; or you want to distribute records with a particular column value within a certain range to target A, within a different range to target B, and within a still different range to target C.
This conditional distribution of source column data is sometimes called row selection or row partitioning, just as the mapping of a selected subset of all the columns is sometimes called column selection or column partitioning.
So far, we had limited our discussion to cases where the values of the mapped columns on all records are copied, or the value changes for the mapped columns on all changed records are replicated. The one exception was data for unmapped columns that is not distributed.