This post is a part of the DP-600: Implementing Analytics Solutions Using Microsoft Fabric Exam Prep Hub; and this topic falls under these sections: 
Prepare data 
    --> Get data 
        --> Create a data connection

Creating data connections is a foundational skill for a Microsoft Fabric Analytics Engineer. In the DP-600 exam, this topic focuses on how to securely and efficiently connect Fabric workloads—such as Lakehouses, Warehouses, Dataflows Gen2, and semantic models—to a wide variety of data sources.

What a Data Connection Means in Microsoft Fabric

A data connection defines how Fabric authenticates to, accesses, and retrieves data from a source system. It includes:

• The data source type
• Connection details (server, database, endpoint, file path, etc.)
• Authentication method
• Optional privacy and credential reuse settings

Once created, a data connection can often be reused across multiple items within a workspace.

Common Data Sources in Fabric

For the exam, you should be familiar with connecting to the following categories of data sources:

1. Azure and Microsoft Data Sources

• Azure SQL Database
• Azure Synapse (dedicated and serverless pools)
• Azure Data Lake Storage Gen2
• Azure Blob Storage
• OneLake (Fabric-native storage)
• Power BI semantic models (DirectQuery)

2. On-Premises Data Sources

• SQL Server
• Oracle
• Other relational databases

These typically require an On-premises Data Gateway.

3. Files and Semi-Structured Data

• CSV, JSON, Parquet, Excel
• Files stored in OneLake, ADLS Gen2, SharePoint, or local file systems

Where Data Connections Are Created

In Microsoft Fabric, data connections can be created from several entry points:

• Lakehouse: Add data via shortcuts or ingestion
• Warehouse: Connect external data or ingest via pipelines
• Dataflows Gen2: Define connections as part of Power Query Online
• Pipelines: Configure source connections in copy activities
• Semantic models: Connect via Import or DirectQuery

Understanding where the connection is configured is important for exam scenarios.

Authentication Methods

The DP-600 exam commonly tests authentication concepts. Be familiar with:

• Microsoft Entra ID (OAuth) – Recommended and most secure
• Service principal – Common for automation and CI/CD
• Account key / Shared Access Signature (SAS) – Often used for storage
• Username and password – Less secure, sometimes legacy

You should also understand when credentials are:

• Stored at the connection level
• Managed per workspace
• Reused across multiple items

Gateways and Connectivity Modes

On-Premises Data Gateway

Required when connecting Fabric to on-premises sources. Key points:

• Can be standard or personal (standard is preferred)
• Must be online for refresh and query operations
• Uses outbound connections only

Connectivity Modes

• Import: Data is loaded into Fabric storage
• DirectQuery: Queries run against the source system
• Shortcut-based access: Data remains external but appears native in OneLake

Security and Governance Considerations

When creating data connections, Fabric enforces governance through:

• Workspace roles (Viewer, Contributor, Member, Admin)
• Credential isolation per workspace
• Sensitivity labels inherited from data sources (when applicable)

Exam questions may test your ability to choose the most secure and scalable connection method.

Best Practices (Exam-Relevant)

• Prefer Entra ID authentication over credentials or keys
• Use OneLake shortcuts to avoid unnecessary data duplication
• Centralize connections in Dataflows Gen2 for reuse
• Validate gateway availability for on-premises sources
• Align connection methods with performance needs (Import vs DirectQuery)

How This Appears on the DP-600 Exam

You may be asked to:

• Identify the correct data connection method for a scenario
• Choose the appropriate authentication type
• Determine when a gateway is required
• Decide where to create a connection for reuse and governance
• Troubleshoot refresh or connectivity issues

Key Takeaway
Creating data connections in Microsoft Fabric is about more than just accessing data—it’s about security, performance, reusability, and governance. For the DP-600 exam, focus on understanding source types, authentication options, gateways, and where connections are defined within the Fabric ecosystem.

Practice Questions:

Here are 10 questions to test and help solidify your learning and knowledge. As you review these and other questions in your preparation, make sure to …

• Identifying and understand why an option is correct (or incorrect) — not just which one
• Look for and understand the usage scenario of keywords in exam questions (for example, gateway, authentication, reuse, DirectQuery vs Import)
• Expect scenario-based questions rather than direct definitions

1. Which authentication method is generally recommended when creating data connections in Microsoft Fabric?

A. Username and password
B. Shared Access Signature (SAS)
C. Microsoft Entra ID (OAuth)
D. Account key

Correct Answer: C

Explanation:
Microsoft Entra ID (OAuth) is the recommended authentication method because it provides centralized identity management, better security, support for conditional access, and easier credential rotation compared to passwords or keys.

2. When is an On-premises Data Gateway required in Microsoft Fabric?

A. When connecting to Azure SQL Database
B. When connecting to OneLake
C. When connecting to an on-premises SQL Server
D. When connecting to Azure Data Lake Storage Gen2

Correct Answer: C

Explanation:
An On-premises Data Gateway is required when Fabric needs to access data sources that are hosted on-premises. Cloud-based sources such as Azure SQL Database or ADLS Gen2 do not require a gateway.

3. Which Fabric feature allows external data to appear as if it is stored in OneLake without copying the data?

A. Import mode
B. DirectQuery mode
C. OneLake shortcuts
D. Data pipelines

Correct Answer: C

Explanation:
OneLake shortcuts provide a logical reference to external storage locations (such as ADLS Gen2 or S3) without physically moving or duplicating the data.

4. You want multiple Fabric items in the same workspace to reuse a single data connection. Where should you create the connection?

A. In each semantic model
B. In Dataflows Gen2
C. In Power BI Desktop only
D. In Excel

Correct Answer: B

Explanation:
Dataflows Gen2 are designed for centralized data ingestion and transformation, making them ideal for creating reusable data connections across multiple Fabric items.

5. Which connectivity mode loads data into Fabric storage and provides the best query performance?

A. DirectQuery
B. Live connection
C. Shortcut-based access
D. Import

Correct Answer: D

Explanation:
Import mode copies data into Fabric-managed storage, enabling high-performance queries and full modeling capabilities at the cost of data freshness.

6. Which statement about DirectQuery connections in Fabric is true?

A. Data is stored in OneLake
B. Queries are always faster than Import mode
C. Queries are executed against the source system
D. A gateway is never required

Correct Answer: C

Explanation:
With DirectQuery, queries are sent directly to the source system at runtime. Performance depends on the source, and a gateway may be required for on-premises sources.

7. Which role is required to create or edit data connections within a Fabric workspace?

A. Viewer
B. Contributor
C. Member
D. Admin

Correct Answer: B

Explanation:
Users must have at least Contributor permissions to create or modify data connections. Viewers have read-only access and cannot manage connections.

8. Which file formats are commonly supported when creating file-based data connections in Fabric?

A. CSV only
B. CSV, JSON, Parquet, Excel
C. TXT only
D. XML only

Correct Answer: B

Explanation:
Microsoft Fabric supports a wide range of structured and semi-structured file formats, including CSV, JSON, Parquet, and Excel, especially when stored in OneLake or ADLS Gen2.

9. What is the primary security benefit of using a service principal for data connections?

A. Faster query performance
B. No need for a gateway
C. Automated, non-interactive authentication
D. Unlimited access to all workspaces

Correct Answer: C

Explanation:
Service principals enable secure, automated authentication scenarios (such as CI/CD pipelines) without relying on individual user credentials.

10. A data refresh in Fabric fails because credentials are missing. What is the most likely cause?

A. The dataset is in Import mode
B. The gateway is offline or misconfigured
C. The semantic model contains calculated columns
D. The file format is unsupported

Correct Answer: B

Explanation:
If a data source requires an On-premises Data Gateway and the gateway is offline or incorrectly configured, Fabric cannot access the credentials, causing refresh failures.