For more information on the formats supported for input, see Supported File Formats.For more information on distributed sources of datasets, see Connection Types.Steps:

When data is imported, a reference to it is stored by the platform as an imported dataset. The source data is not modified. In the application, you modify the recipe associated with a dataset to transform the imported data.

1. Login to the application.

   Tip
   Tip: When you login for the first time, you can immediately import a dataset to begin transforming it.

2. In the menubar, click Datasets Library. Click Import Data.

3. To add a dataset:

   1. Select the connection where your source is located. For this basic workflow, select Upload to upload a file from your local desktop. You can select multiple files to upload. For this example, select only one file.
   2. Navigate and select the file or files for your source. Click Open.
   3. To queue the dataset for uploading, click the Plus icon next to its name.

4. To begin working with a your dataset, you must first add it into a flow, which is a container for datasets. Click the Add Dataset to a Flow checkbox and enter the name for a new flow. Click Import & Add to Flow.

   TipTip: If you have selected a single file, you can begin wrangling it immediately. Click Import and Wrangle. The

   click Continue. The flow is created for you

   and is named based on the dataset name. Your

   , and your dataset is automatically added to it.

5. After the flow has been created, you You can begin working with the flow objects. See Flow View Page.
6. Select the dataset and click Create New Recipe. 
7. Now, click Edit Recipe to begin working immediately with the dataset in the Transformer page. See Transform Basics.

For  For more information, see Import Tasks.

...

http://trifacta-public-datasets.s3.amazonaws.com/demo-datasets/customers_sample_data.csvThe above dataset is used as part of the on-boarding tour presented to new users.