Get started with Databricks administration

This article article provides opinionated guidance for new account and workspace admins looking to take advantage of the administrative and security features available on Databricks. For more in-depth security guidance, see the Security and compliance guide.

Requirements

You need a Databricks account and workspace. If you haven’t set yours up yet, follow the steps in Get started: Account and workspace setup to get up and running. Once you have a workspace set up, go through the following admin tasks:

Step 1: Build out your team

The best practice for building out your team is to add users and groups to your account by syncing your identity provider (IdP) with Databricks. If you choose to build your team out manually, you can follow the steps in Manage users and Manage groups to add your team through the account console UI.

You should organize your users and service principals into account groups based on permissions and roles. Account groups simplify identity management by making it easier to assign access to workspaces, data, and other securable objects.

After your team has been added to Databricks, the following tasks are recommended:

• Assign groups to the workspace

• Assign the account admin role to a group

Step 2: Configure permissions and access control

Within a workspace, workspace admins help secure data and control compute usage by giving users access only to the Databricks functionality and data they need.

Note

Access control requires the Premium plan. If you don’t have it, go to the account console to update your subscription or contact your Databricks account team.

The following articles walk you through enabling and managing key features workspace admins can use to control data access and compute usage:

• Manage data governance and user data access

• Manage access control for clusters, jobs, notebooks, and other workspace objects

• Create and manage compute policies

Step 3: Set up account monitoring

To control costs and allow your organization to monitor detailed Databricks usage patterns, including audit and billable usage logs, Databricks recommends using system tables (Public Preview). You can also use custom tags to help monitor resources and data objects.

• Monitor usage with system tables.

• Monitor usage using tags

Step 4: Implement additional security features

Databricks provides a secure networking environment by default, but if your organization has additional needs, you can configure network security features on your Databricks resources. See Customize network security. For an overview of available security features, see Security and compliance guide.

Get Databricks support

If you have any questions about setting up Databricks and need live help, please e-mail onboarding-help@databricks.com.

If you have a Databricks support package, you can open and manage support cases with Databricks. See Learn how to use Databricks support.

If your organization does not have a Databricks support subscription, or if you are not an authorized contact for your company’s support subscription, you can get answers to many questions in Databricks Office Hours or from the Databricks Community.

Databricks Academy

Databricks Academy has a free self-paced learning path for platform administrators. Before you can access the course, you first need to register for Databricks Academy if you haven’t already.

You can also sign up to attend a live platform administration training.

Additional resources

The following table includes links for further learning:

Become a Databricks expert	Run the Get started quickstarts Take training courses at Databricks Academy Review resources such as e-books, webinars, and more
Learn industry best practices and news	Read Databricks blogs View past Spark + AI Summit sessions
Follow in-depth, proven best practices	Learn about CI/CD on Databricks Learn how to keep your data lake GDPR and CCPA compliant using Delta Lake Meet HIPAA requirements Manage data retention Learn how to migrate workloads to Databricks Learn cluster configuration best practices on Databricks
Get involved	Participate in Data + AI Summit Influence the product roadmap by adding ideas to the Ideas Portal
Get help and support	Help Center Community Forums