The Data Integration Service applies optimization methods in an attempt to reduce the amount of data to process. When you run a mapping, you can choose an optimizer level that determines which optimization methods the Data Integration Service can apply to the mapping. For example, the Data Integration Service can use early selection optimization to move a filter closer to the source. It can use pushdown optimization to push transformation logic to a database. It can use the cost-based optimization method to change the join processing order.

The Data Integration Service can apply multiple optimization methods to a mapping at the same time. For example, the Data Integration Service applies the early projection, predicate optimization, early selection, branch pruning, or push-into optimization methods when you select the normal optimizer level.

If you have the partitioning option, the Data Integration Service can maximize parallelism for mappings. The Data Integration Service dynamically determines the partitioning strategy for mappings. The partitioning strategy includes the location of partition points, the optimal number of partitions for each pipeline stage, and the partitioning types that best redistribute data across each partition point. For more information about partitioning, see Partitioned Mappings Overview.