Run federated queries on Microsoft SQL Server

This article describes how to set up Lakehouse Federation to run federated queries on SQL Server data that is not managed by Azure Databricks. To learn more about Lakehouse Federation, see What is Lakehouse Federation?.

To connect to your SQL Server database using Lakehouse Federation, you must create the following in your Azure Databricks Unity Catalog metastore:

• A connection to your SQL Server database.
• A foreign catalog that mirrors your SQL Server database in Unity Catalog so that you can use Unity Catalog query syntax and data governance tools to manage Azure Databricks user access to the database.

Lakehouse Federation supports SQL Server, Azure SQL Database, and Azure SQL Managed Instance.

Before you begin

Workspace requirements:

• Workspace enabled for Unity Catalog.

Compute requirements:

• Network connectivity from your compute resource to the target database systems. See Networking recommendations for Lakehouse Federation.
• Azure Databricks compute must use Databricks Runtime 13.3 LTS or above and Shared or Single user access mode.
• SQL warehouses must be pro and must use 2023.40 or above.

Permissions required:

• To create a connection, you must be a metastore admin or a user with the CREATE CONNECTION privilege on the Unity Catalog metastore attached to the workspace.
• To create a foreign catalog, you must have the CREATE CATALOG permission on the metastore and be either the owner of the connection or have the CREATE FOREIGN CATALOG privilege on the connection.

Additional permission requirements are specified in each task-based section that follows.

• If you plan to authenticate using OAuth, register an app in Microsoft Entra ID for Azure Databricks. See the following section for details.

(Optional) Register an app in Microsoft Entra ID for Azure Databricks

If you want to authenticate using OAuth, follow this step before you create a SQL Server connection. To authenticate using a username and password instead, skip this section.

1. Sign in to the Azure portal.
2. In the left navigation, click Microsoft Entra ID.
3. Click App registrations.
4. Click New registration. Enter a name for the new app and set the redirect URI to https://<workspace-url>/login/oauth/azure.html.
5. Click Register.
6. In the Essentials box, copy and store the Application (client) ID. You'll use this value to configure the application.
7. Click Certificates & secrets.
8. On the Client secrets tab, click New client secret.
9. Enter a description for the secret and an expiration (the default setting is 180 days).
10. Click Add.
11. Copy the generated value for the client secret.
12. Click API permissions.
13. Click Add a permission.
14. Select Azure SQL Database and click user_impersonation under Delegated permissions.
15. Click Add permissions.

Create a connection

A connection specifies a path and credentials for accessing an external database system. To create a connection, you can use Catalog Explorer or the CREATE CONNECTION SQL command in an Azure Databricks notebook or the Databricks SQL query editor.

Permissions required: Metastore admin or user with the CREATE CONNECTION privilege.

Catalog Explorer

1. In your Azure Databricks workspace, click Catalog.

2. At the top of the Catalog pane, click the Add icon and select Add a connection from the menu.

   Alternatively, from the Quick access page, click the External data button, go to the Connections tab, and click Create connection.

3. Enter a user-friendly Connection name.

4. Select a Connection type of SQL Server.

5. Select an Auth type of OAuth or Username and password.

6. Enter the following connection properties for your SQL Server instance, depending on your authentication method:

   • Host: Your SQL server.
   • (Basic authentication) Port
   • (Basic authentication) trustServerCertificate: Defaults to false. When set to true, the transport layer uses SSL to encrypt the channel and bypasses the certificate chain to validate trust. Leave this set to the default unless you have a specific need to bypass trust validation.
   • (Basic authentication) User
   • (Basic authentication) Password
   • (OAuth) Authorization Endpoint: Your Azure Entra authorization endpoint in the format https://login.chinacloudapi.cn/<tenant-id>/oauth2/v2.0/authorize.
   • (OAuth) Client ID from the app you created.
   • (OAuth) Client secret from the client secret you created.
   • (OAuth) Client scope: Enter the following value with no modifications: https://database.chinacloudapi.cn/.default offline_access.
   • (OAuth) You're prompted to sign in to Log in with Azure Entra ID. Enter your Azure username and password. After you're redirected to the connection creation page, the authorization code is populated in the UI.

7. (Optional) Click Test connection to confirm that it works.

8. (Optional) Add a comment.

9. Click Create.

SQL

Run the following command in a notebook or the Databricks SQL query editor.

CREATE CONNECTION <connection-name> TYPE sqlserver
OPTIONS (
  host '<hostname>',
  port '<port>',
  user '<user>',
  password '<password>'
);

We recommend that you use Azure Databricks secrets instead of plaintext strings for sensitive values like credentials. For example:

CREATE CONNECTION <connection-name> TYPE sqlserver
OPTIONS (
  host '<hostname>',
  port '<port>',
  user secret ('<secret-scope>','<secret-key-user>'),
  password secret ('<secret-scope>','<secret-key-password>')
)

For information about setting up secrets, see Secret management.

Create a foreign catalog

A foreign catalog mirrors a database in an external data system so that you can query and manage access to data in that database using Azure Databricks and Unity Catalog. To create a foreign catalog, you use a connection to the data source that has already been defined.

To create a foreign catalog, you can use Catalog Explorer or the CREATE FOREIGN CATALOG SQL command in an Azure Databricks notebook or the SQL query editor.

Permissions required: CREATE CATALOG permission on the metastore and either ownership of the connection or the CREATE FOREIGN CATALOG privilege on the connection.

Catalog Explorer

1. In your Azure Databricks workspace, click Catalog to open Catalog Explorer.

2. At the top of the Catalog pane, click the Add icon and select Add a catalog from the menu.

   Alternatively, from the Quick access page, click the Catalogs button, and then click the Create catalog button.

3. Follow the instructions for creating foreign catalogs in Create catalogs.

SQL

Run the following SQL command in a notebook or SQL query editor. Items in brackets are optional. Replace the placeholder values:

• <catalog-name>: Name for the catalog in Azure Databricks.
• <connection-name>: The connection object that specifies the data source, path, and access credentials.
• <database-name>: Name of the database you want to mirror as a catalog in Azure Databricks.

CREATE FOREIGN CATALOG [IF NOT EXISTS] <catalog-name> USING CONNECTION <connection-name>
OPTIONS (database '<database-name>');

Supported pushdowns

The following pushdowns are supported on all compute:

• Filters
• Projections
• Limit
• Functions: partial, only for filter expressions. (String functions, Mathematical functions, Data, Time and Timestamp functions, and other miscellaneous functions, such as Alias, Cast, SortOrder)

The following pushdowns are supported on Databricks Runtime 13.3 LTS and above, and on SQL warehouse compute:

• Aggregates
• The following Boolean operators: =, <, <=, >, >=, <=>
• The following mathematical functions (not supported if ANSI is disabled): +, -, *, %, /
• The following miscellaneous operators: ^, |, ~
• Sorting, when used with limit

The following pushdowns are not supported:

• Joins
• Windows functions

Data type mappings

When you read from SQL Server to Spark, data types map as follows:

SQL Server type	Spark type
bigint (unsigned), decimal, money, numeric, smallmoney	DecimalType
smallint, tinyint	ShortType
int	IntegerType
bigint (if signed)	LongType
real	FloatType
float	DoubleType
char, nchar, uniqueidentifier	CharType
nvarchar, varchar	VarcharType
text, xml	StringType
binary, geography, geometry, image, timestamp, udt, varbinary	BinaryType
bit	BooleanType
date	DateType
datetime, datetime, smalldatetime, time	TimestampType/TimestampNTZType

*When you read from SQL Server, SQL Server datetimes are mapped to Spark TimestampType if preferTimestampNTZ = false (default). SQL Server datetimes are mapped to TimestampNTZType if preferTimestampNTZ = true.