The most common performance bottleneck occurs when the Data Integration Service writes to a target database. If the database uses small checkpoint intervals, the database processing slows more often write a checkpoint. Small database network packet sizes can cause bottlenecks. You can allow larger packets of data to cross the network at one time.

To identify a target bottleneck, you can create a copy of the mapping that has a flat file target instead of a database target. If thes performance increases significantly, you have a target bottleneck. If the mapping already writes to a flat file target, you probably do not have a target bottleneck.