Creating datasets using the Empty Source task

You can create new datasets from scratch, using the Empty Source task, simply by specifying the size of the dataset.

Tasks such as these are useful when you want to hard-code data into a process, for example when mapping values, independent from external sources. The values themselves can be inserted via a subsequent Data Manager task.

These tasks can also be useful when prioritizing tasks, and you want to add a sentry task that will always be run, without generating errors.


  • You must have created a flow;


  1. Drag and drop the Empty Source task onto the central stage.

  2. Double click the task.

  3. In the Number of attributes box (natt) enter how many attributes will be included in your dataset. 

  4. In the Number of records box (ndata) enter how many records will be included in your dataset.

  5. Save and compute the task.