Navigating Dataflows: The Mechanics

Objective:

To utilize the power of the Dataflows tool, you first need to know some of the basic mechanics. This tutorial is designed to help acquaint you with the functionality of the interface. Some of the fundamental skills you will learn in this tutorial include how to:

  • Create a dataflow
  • Import and Load a dataset
  • Store a dataset
  • Save a dataflow
  • Open and modify an existing dataflow

Prerequisites:

Prior to beginning this tutorial, we have already logged in to the Aunsight platform and navigated to our organization. Since we will be importing, loading, and storing a dataset in this tutorial, we have already created a dataset. If you would like to review these steps, please visit the following tutorials.

Login to Aunsight

Create a Dataset

How to Create a Dataflow

1. Select the Dataflows Tool

In the panel on the left, select Dataflows. Select Dataflows

Get to Know the Dataflows Interface:

In the Dataflows interface, there is a panel that lists existing dataflows, if any, for the selected context. There are tools available for sorting and searching existing dataflows. Pagination tools are available at the bottom of the existing dataflow list to browse other pages. If an existing dataflow is selected, the details will appear in the main pane of the page.

Dataflows Interface Overview

2. Choose the "+" Icon to Add a New Dataflow

Select the "+" icon at the top of the list of existing dataflows.
Select Add Icon

The window will change to the Dataflow Editor. Dataflow Editor

Get to Know the Dataflows Editor:

There are four primary components of the Dataflow Editor. The left panel navigates imported datasets. The next panel displays operation details. The main panel acts as a canvas to display the relationship between different operations in a graphical layout. Finally, the toolbar across the top provides quick access buttons for common tasks and the operation selector.

DataflowsEditor

3. Import a Dataset

Select the Dataset Icon to import an existing dataset. When the prompt appears, choose the dataset to import. If you need to, use the search option to locate your dataset. Before you can perform operations on your dataset later, it will need to be loaded into the dataflow as well. The option to import and load the dataset is chosen by default. Choose Import. The selected dataset should appear in the Resource Panel at the left of the screen and in a tile on the canvas.

ImportLoadDataset

Get to Know the Interface: Dataset and Operation Details

In the Resources Panel, when the Bulleted List Icon is selected, the fields in the dataset are displayed. When the Table Icon is selected, the list of imported datasets is displayed.

In the Canvas, when the Load Operation is selected, the details are displayed in the Properties Panel.

DatsetOperationDetails

4. Store a Dataset

To store a dataset at the end of a series of operations, select the "+" icon to add a new operation. When the menu appears, select "Store" from the list. The Properties Panel will automatically shift to display the details pertinent to the new store operation.

Complete the details for the store operation by typing a title and description. Choose the appropriate input(s), if they have not been selected automatically, by clicking the input(s) field and then the appropriate dataset name from the Canvas. Choose the output dataset. The choices will include a new dataset or any datasets that have been imported but not loaded in the dataflow. Choose apply.

Note

Be mindful about overwriting existing datasets. To learn more about the process of using the store operation to overwrite existing datasets and the appropriate use case, visit the following tutorial. Overwriting Datasets

After applying the changes to the Store operation, the Canvas and Properties Panel will update to reflect the added operation.

Store a Dataset

How to Save a Dataflow

To save a dataflow, select the "Save" button on the right side of the toolbar. When the prompt opens, type the intended name of the dataflow. While the name field is required, there is a text-entry field to provide an optional description. If appropriate, choose the related tags. Submit the changes.

The breadcrumb list under the organization context will update to include the new name of the dataflow. This confirms that the save was successful.

Save a Dataflow

How to Open and Modify an Existing Dataflow

1. Select the Existing Dataflow

From the list of existing dataflows, select the name of the intended one. Each time the name of a dataflow is selected on the list, the main pane will update to display the details of that dataflow.

Select a Dataflow

2. Select Modify

To open an existing dataflow in the Dataflow Editor, select Modify. Dataflow Modify

Warning

Don't forget to save your changes as you modify your dataflow.

Next Steps:

Now that you have created a new dataflow and saved it, you might want to take a look at how some common operations are implemented. To complete your first dataflow with a common sequence of operations, visit the following tutorial. Use Case 1