In round-robin partitioning, the PowerCenter Integration Service distributes blocks of data to one or more partitions. Each partition processes rows based on the number and size of the blocks.
Use round-robin partitioning when you do not need to group data among partitions. In a pipeline that reads data from file sources of different sizes, use round-robin partitioning to distribute blocks of rows between the partitions.
The following figure shows a mapping where round-robin partitioning helps distribute rows before they enter a Filter transformation:
The session based on this mapping reads item information from three flat files of different sizes:
Source file 1: 80,000 rows
Source file 2: 5,000 rows
Source file 3: 15,000 rows
When the PowerCenter Integration Service reads the source data, the first partition begins processing 80% of the data, the second partition processes 5% of the data, and the third partition processes 15% of the data.
To distribute the workload more evenly, set a partition point at the Filter transformation and set the partition type to round-robin. The PowerCenter Integration Service distributes the data so that each partition processes approximately one-third of the data.