DP-600 Lab Summary Series - Lab 1 Create a Microsoft Fabric Lakehouse

This article is part of the "DP-600 Lab Summary Series" that summarises the key lessons of each of Microsoft's Lab exercises as well as explaining the code (where relevant).

The aim is use these as study notes and refresher for the DP-600 exams especially for those who are beginners or newbies to Microsoft Fabric (just like me).

Link to this Lab Exercise: https://microsoftlearning.github.io/mslearn-fabric/Instructions/Labs/01-lakehouse.html

I suggest you first complete the Lab exercise and then come back to this article for the study notes.

Key lessons

Data Warehousing vs. Data Lake:

• Data Warehousing: Uses relational tables with a fixed schema for storage and querying using SQL.
• Data Lake: Stores data as files without a fixed schema, enabling storage of unstructured, semi-structured, and structured data.
• Data Lakehouse: Combines the strengths of both approaches by storing data in files in a data lake and applying a relational schema as a metadata layer for querying with SQL.

Microsoft Fabric:

• Provides a data lakehouse solution with OneLake store for scalable file storage and Delta Lake for managing relational objects.
• Delta Lake enables defining schema on files in the lakehouse, ensuring consistency and enabling efficient querying.

Workspace Creation:

• Creating a workspace is the first step in using Fabric, providing a dedicated environment for your data engineering tasks.
• The workspace can be configured with different licensing modes, including a trial mode with Fabric capacity.

Lakehouse Creation:

• Within the workspace, you can create a data lakehouse, which serves as the central storage and query layer for your data.
• The lakehouse is based on the OneLake store and includes a metastore for managing tables and views.

Data Upload:

• Data can be uploaded into the lakehouse using various methods, including direct upload, pipelines, and data flows.
• Shortcuts can be created to reference externally stored data without physically moving it into the lakehouse.

Table Creation:

• While data in files can be queried directly using Apache Spark, creating tables enables querying the data using SQL.
• Tables are based on the Delta Lake format, providing transactional capabilities and schema enforcement.

Querying with SQL:

• A SQL endpoint is automatically created for the lakehouse, allowing you to query tables using SQL SELECT statements.
• Queries can be executed to retrieve and analyze data stored in the lakehouse, leveraging traditional SQL semantics.

Visual Querying:

• Data analysts familiar with Power BI can use Power Query skills to create visual queries and reports based on the tables in the lakehouse.
• Visual queries enable interactive exploration and visualization of data, enhancing data analysis capabilities.

Report Creation:

• Tables in the lakehouse are automatically included in a default semantic model, which forms the basis for reporting with Power BI.
• Reports can be created using the Power BI interface, leveraging the data model to visualize and analyze data from the lakehouse.

Resource Cleanup:

• Once you have finished exploring your lakehouse, you can delete the workspace to clean up resources and stop incurring costs.
• Deleting the workspace removes all associated resources, including the lakehouse, SQL endpoint, and any data models or reports created within the workspace.

Hope you found this useful.

Be on the lookout for the next article in this series.

Until next time.

Arno Wakfer

Disclaimer: The summaries in this article were provided with the assistance of AI. AI is not perfect and it can make mistakes. Its always recommended to double check the content with reliable resources.

#DP600

#MicrosoftFabric