---
title: Data connections
description: Connect to an external data source to seamlessly browse, preview, and profile data, as well as initiate scalable data preparation for machine learning with push-down.

---

# Data connections {: #data-connections }

In Workbench, you can easily configure and reuse secure connections to predefined data sources. Not only does this allow you to interactively browse, preview, and profile your data, it also gives you access to DataRobot's integrated [data preparation capabilities](wb-wrangle-data/index).

See the associated [considerations](wb-data-ref/index#add-data){ target=_blank } for important additional information.

??? warning "Source IP addresses for allowing"
    Before setting up a data connection, make sure the [source IPs](data-conn#allowed-source-ip-addresses){ target=_blank } have been allowed.

{% include 'includes/workbench/wb-pp-dynamic.md' %}

## Supported connections {: #supported-connections }

Workbench currently supports the following connections:

Connection  |  Notes
---------- | ------------------
Snowflake | See the [documentation](dc-snowflake){ target=_blank } for required parameters and additional information.
BigQuery | See the [documentation](dc-bigquery){ target=_blank } for required parameters and additional information.
Databricks<br>(public preview) | See the [documentation](wb-databricks){ target=_blank } for required parameters and additional information.
S3<br>(public preview) | See the [documentation](dc-s3){ target=_blank } for required parameters and additional information.


!!! info "Public preview"
    Support for Databricks in Workbench is on by default.

    **Feature flag(s):** Enable Databricks Driver


!!! info "Public preview"
    Support for AWS S3 in Workbench is on by default.

    **Feature flag(s):** Enable Native S3 Driver

For a complete list of available connections in Workbench and which features they support, see the [connection capabilities table](wb-data-ref/index#connection-capabilities).

## Connect to a data source {: #connect-to-a-data-source }

Creating a data connection lets you explore external source data and then add it to your Use Case.

To create a data connection:

1. In a Use Case, click **Add new > Add datasets**. The **Add data** modal opens.

    ![](images/wb-databricks-1.png)

2. Click **Connect**.

    ![](images/wb-connect-1.png)

3. Select the data source (Snowflake in this example).

    ![](images/wb-connect-2.png)

    Now, you can [configure the data connection](#configure-the-connection).

## Configure the connection {: #configure-the-connection }

!!! note
     When configuring your data connection, configuration types, authentication options, and required parameters are based on the selected data source. The example below shows how to configure Snowflake with OAuth using new credentials.

To configure the data connection:

1. On the **Configuration** page, select a configuration method&mdash;either **Parameters** or **JDBC URL**.

2. Enter the required parameters for the selected configuration method.

    === "Parameters"

        ![](images/wb-connect-7.png)

    === "JDBC URL"

        ![](images/wb-connect-6.png)

3. Click **New Credentials** and select an authentication method&mdash;in this case, either **Basic** or **OAuth**.

    === "Basic"

        ![](images/wb-connect-8.png)

    === "OAuth"

        ![](images/wb-connect-9.png)

    ??? note "Saved credentials"
        If you previously [saved credentials for the selected data source](stored-creds#credentials-management){ target=_blank }, click **Saved credentials** and select the appropriate credentials from the dropdown.

        ![](images/wb-connect-10.png)

4. Click **Save** in the upper right corner.

    ![](images/wb-connect-3.png)

     If you selected OAuth as your authentication method, you will be prompted to sign in before you can [select a dataset](#select-a-dataset). See the [DataRobot Classic documentation](dc-snowflake){ target=_blank } for more information about supported authentication methods and required parameters.

## Select a dataset {: #select-a-dataset }

Once you've set up a data connection, you can add datasets by browsing the [database schemas](https://www.ibm.com/topics/database-schema){ target=_blank } and tables you have access to.

To select a dataset:

1. Select the schema associated with the table you want to add.

    ![](images/wb-connect-4.png)

2. Select the box to the left of the appropriate table.

    ![](images/wb-connect-5.png)

    With a dataset selected, you can:

    |          | Description |
    |----------|-------------|
    | <div class="table-label">1</div> | Click **Wrangle** to prepare the dataset before adding it to your Use Case. |
    | <div class="table-label">2</div> | Click **Preview** to open a snapshot preview to help determine if the dataset is relevant to your Use Case and/or if it needs to be wrangled. |
    | <div class="table-label">3</div> | Click **Add to Use Case** to add it to your Use Case, making it available to you and other team members on the Datasets tab. |

    ??? tip "Large datasets"
        If you want to decrease the size of the dataset before adding it to your Use Case, click Wrangle. When you publish a recipe, you can [configure automatic downsampling](wb-pub-recipe#configure-downsampling){ target=_blank } to control the number of rows when Snowflake materializes the output dataset.

## Next steps {: #next-steps }

From here, you can:

- [Perform data wrangling before adding the dataset to your Use Case.](wb-wrangle-data/wb-add-operation)
- [Add more data.](wb-add-data/index)
- [View exploratory data insights for the dataset.](wb-data-tab)
- [Use the dataset to set up an experiment and start modeling.](wb-experiment/index)

## Read more {: #read-more }

To learn more about the topics discussed on this page, see:

- [Workbench connection capabilities.](wb-data-ref/index#connection-capabilities){ target=_blank }
- [DataRobot's dataset requirements.](file-types){ target=_blank }
- [Saved data connection credentials.](stored-creds){ target=_blank }
- [Delete data connections.](data-conn#delete-a-connection){ target=_blank }
- [Workbench allowed source IPs.](data-conn#allowed-source-ip-addresses){ target=_blank }
