When you add a Human task to a workflow, you configure one or more instances of the task. You assign users to the task instances, so that each user works on part of the data set. The Data Integration Service creates the task instances when the workflow runs.
Use the task distribution options in the Human task to create multiple instances of the task. The task distribution options determine the size and composition of the data in each task instance. For example, if the input data set for the Human task contains one thousand records, you can create five task instances with two hundred records in each instance.
You can create task instances in the following ways:
Specify the number of task instances to create. The Data Integration Service divides the input data set into task instances of equal size.
Specify the number of rows or clusters to include in a task. The Data Integration Service creates task instances that contain the number of rows or clusters you specify. The number of task instances depends on the number of rows or clusters in the input data set.
Specify the data rows or clusters to assign to each user. You select an input data column, and you identify the users who can work on each task instance. The Data Integration Service assigns rows or clusters to users based on the values in the column you select.
If you add multiple steps to a Human task, the data associated with each task instance passes through every step in the Human task. You select the users who work on the task instance in each step. You can assign the same set of users to all steps in the Human task, or you can assign different users to each step.