Set up your Databricks account and deploy a workspace

To sign up for a free trial of Databricks, click the TRY DATABRICKS button at the top of this page.

Once you have signed up for a Databricks account, it’s time to set it up and create at least one workspace, which is the environment that your team will use for accessing all of their Databricks assets.

As the user who signed up for Databricks, you are the account owner, and only you can perform initial setup. However, you can assign other users as account admins to perform follow-on account administration tasks.

Note

The instructions that follow explain how to use the Databricks account console and AWS user interfaces to set up your Databricks account and create a workspace manually. As an alternative, you can use the Databricks on AWS Quickstarts, which take advantage of CloudFormation templates to configure your account and workspaces more quickly and programmatically. These quickstarts are targeted to IT infrastructure architects, administrators, and DevOps professionals.

Initial setup

1. Open the Welcome email you received from Databricks after you signed up for an account and click the link to verify your email address and create your password.

   When you click Submit, you are taken directly to the Databricks account console.

2. Select a subscription plan.

   

   For information about subscription plans and pricing, see the Databricks AWS pricing page.

   Note

   If a Databricks representative has assisted you with your Databricks subscription, you don’t need to make any selections on this page.

3. Click Continue to open the Workspaces page.

4. Click Create workspace to set up a Databricks workspace.

   A workspace is the environment that your team members use for accessing all of their Databricks assets.

5. On the Create Workspace page, add a workspace name and region, and confirm the subscription plan:

   • Workspace name: Human readable name for your workspace. For example, “Acme data science workspace.”
   • Subscription plan: The plan that you selected on the subscription page.
   • Region: The AWS region containing your Databricks workspace’s resources.

6. Create a credential configuration.

   A credential configuration provides information about the cross-account IAM role you must create to give Databricks limited access to launch clusters in your AWS account.

   To create the IAM role and credential configuration:

   1. In the Credential configuration field, select Add a new credential.

   2. On the Add Credential Configuration dialog, copy the External ID. You will use it when you create the IAM role in the next step.

   3. Follow the instructions in Create a cross-account IAM role.

      Note

      These instructions give you three role policy options, which depend on whether you want to use the default Databricks-managed VPC or provide your own VPC. The typical deployment uses the Databricks-managed VPC, as do all Standard subscription deployments. If you are on a Premium or Enterprise subscription, this is a decision you must make when you create the workspace. You cannot migrate a workspace from a Databricks-managed VPC to a customer-managed VPC.

   4. When you have finished creating the IAM role, copy the IAM role ARN.

   5. Return to the Add Credential Configuration dialog, and paste the role ARN into the Role ARN field.

   6. In Credential Configuration Name, enter a configuration name. Use a name that will help you distinguish this credential configuration from any others that you or your team might create in the future.

   7. Click Add.

   To learn more, see Create a credential configuration.

7. Create a storage configuration.

   A storage configuration provides information about the S3 bucket you must create to store certain Databricks objects (such as cluster logs, notebook revisions, and job results libraries) in your own AWS account.

   To create the S3 bucket and configuration:

   1. In the Storage configuration field, select Add a new storage configuration.
   2. On the Add Storage Configuration dialog, enter a storage configuration name. Use a name that will help you distinguish this storage configuration from any others that you or your team might create in the future.
   3. Enter a bucket name. Use a name that will help you distinguish this S3 bucket from others in your AWS account. You must use this exact name when you create the S3 bucket.
   4. Click Generate Policy to generate a bucket policy that you will use when you create the bucket in your AWS account.
   5. Copy the generated bucket policy and use it when you create the bucket, following the instructions in Create the S3 bucket.

   To learn more, see Manage storage configurations using the account console (E2).

8. (Optional) Click Advanced Settings to create a network configuration to use your own VPC instead of the Databricks default.

   This option is available only on Premium and Enterprise subscriptions.

   By default, Databricks creates a VPC for each workspace it creates in your AWS account. You may instead want to use a customer-managed VPC so that you can configure it according to your organization’s enterprise cloud standards while still conforming to Databricks requirements. To learn more, see Customer-managed VPC.

   To create a network configuration:

   1. Create a VPC that meets the requirements enumerated in Customer-managed VPC. Copy the VPC ID, subnet IDs, and security group IDs.
   2. In the Advanced settings > Network configuration field, select Add a new network configuration.
   3. On the Add Network Configuration dialog, enter a network configuration name. Use a name that will help you distinguish this network configuration from any others that you or your team might create in the future.
   4. Enter the VPC ID, two subnet IDs, and security group IDs you copied when you created the VPC in your AWS account.
   5. Click Add.

9. When your workspace configuration is complete, click Save.

   You should be taken directly to your new workspace. If not, just click the workspace name on the Workspaces page in the account console to take you there. You can also bookmark the workspace URL and log in using the same username (email) and password you used to access the account console.

   

Next steps

Learn how to use Databricks

• Try out the Get started as a Databricks Data Science & Engineering user tutorial.
• Follow the Getting started learning paths and watch the introductory videos.
• Learn Databricks Data Science & Engineering concepts.
• Run the quickstart notebook available on your new workspace’s Welcome page.
• Access free customer training (paid customers only).

Add users, security options, and other administrative tasks

• To delegate account management tasks, you can add account admins. See Delegate account administration.
• To add workspace users, set up security, configure logging, and perform other workspace configurations, see Get started as a Databricks administrator.
• The Administration guide covers all account management and workspace administration tasks.

Add billing information

If you signed up for a 14-day free trial, you can continue using your Databricks account when the trial is over by adding billing information.

1. Log in to the account console as the account owner or an account admin.
2. Click the Settings icon in the sidebar and click the Subscription & Billing tab.
3. Click the Add billing information button.
4. On the Billing page, add your billing information and click Save.

You will be billed monthly until you cancel.

To switch from monthly credit-card billing to invoice or commit billing, contact a Databricks representative.

Learn more about account management

To learn how to manage your Databricks subscription, update account configuration, and more, see Manage your Databricks account (E2).