Discover and connect to data in OneLake

In modern analytics organizations, data often exists across multiple workspaces and teams. Data engineers create lakehouses with raw and cleansed data, other teams build warehouses with business metrics, and analysts develop semantic models. As an analytics engineer, you need to find and connect to this data before you can transform it or build analytical solutions.

Microsoft Fabric provides the OneLake catalog as a centralized discovery experience. The catalog helps you search for data assets, view their metadata, and understand what’s available across your organization. Once you locate the data you need, you can connect to it using shortcuts, query it with SQL, or explore it through semantic models.

In this exercise, you create a sample environment with a lakehouse containing sales data. You then practice discovering this data through the OneLake catalog, creating shortcuts for cross-workspace access, querying data via the SQL analytics endpoint, and exploring semantic models.

This lab takes approximately 30 minutes to complete.

Note: You need access to a Fabric-enabled workspace. If you don’t have one, create a Microsoft Fabric trial to complete this exercise.

Create a workspace

Before working with data in Fabric, create a workspace with the Fabric trial enabled.

  1. Navigate to the Microsoft Fabric home page at https://app.fabric.microsoft.com/home?experience=fabric in a browser, and sign in with your Fabric credentials.

  2. In the menu bar on the left, select Workspaces (the icon looks similar to 🗇).

  3. Create a new workspace with a name of your choice (for example, Data-Engineering).

  4. Choose a Fabric and Power BI workspace type in the Advanced section. The choices might be: Fabric, Fabric trial, Power BI Premium.

  5. When your new workspace opens, it should be empty.

    Screenshot of an empty workspace in Fabric.

Create a lakehouse and load sample data

Now that you have a workspace, it’s time to create a lakehouse and populate it with sample sales data. This lakehouse represents data that a data engineering team might have already created in a real organization.

  1. Download the sales.csv file from https://raw.githubusercontent.com/MicrosoftLearning/dp-data/main/sales.csv, saving it as sales.csv on your local computer (or provided VM, if applicable).

    Note: To download the file, open a new tab in the browser and paste in the URL. Right-click anywhere on the page and select Save as to save it as a CSV file.

  2. In your workspace, select + New item, and then select Lakehouse. Give it a name of your choice (for example, sales-data).

    After a minute or so, a new lakehouse with empty Tables and Files folders will be created.

    Screenshot of a new lakehouse.

  3. In the lakehouse explorer, highlight the Files folder and select the ellipses (…) then Upload > Upload files, and select the sales.csv file you just downloaded.

  4. After the file uploads, select the Files folder to verify that sales.csv was uploaded. Select the file to preview its contents.

    Screenshot of uploaded sales.csv file in a lakehouse.

Load data into a table

Loading file data into a Delta table makes it queryable using SQL and provides reliability features like ACID transactions, which are similar to a traditional data warehouse.

  1. In the Explorer pane, expand the Files folder to see the sales.csv file.

  2. In the ellipses (…) menu for the sales.csv file, select Load to Tables > New table.

  3. In the Load to table dialog, set the table name to sales and confirm the load operation. Wait for the table to be created and loaded.

    Tip: If the sales table doesn’t automatically appear, select Refresh in the ellipses (…) menu for the Tables folder.

  4. In the Explorer pane, select the sales table to view its data and schema.

    Screenshot of a table preview.

Browse the OneLake catalog to discover data

Now that you have data in your lakehouse, you can explore how to discover it through the OneLake catalog. The catalog provides a centralized view of all data assets across your Fabric tenant.

  1. At the top of the page, select the Fabric icon to return to the Fabric home page.

  2. In the left navigation pane, select the OneLake catalog icon.

    Screenshot showing the OneLake catalog with various data assets listed.

  3. In the catalog, you see various item types including lakehouses, warehouses, semantic models, and reports. Browse the list to find your sales_data lakehouse (or whatever name you chose).

    Note: Depending on what else is in your trial environment, you may see other items in the catalog. The catalog respects access permissions, so you only see items you have permission to access.

  4. Select your lakehouse to view its details. The detail pane shows metadata including:
    • Location: The workspace where the item is located
    • Data updated: When the data was last updated
    • Owner: Who created the item
    • SQL connection string: How to connect using tools, such as SQL Server Management Studio (SSMS)

    Screenshot showing lakehouse details in the catalog with metadata displayed.

  5. Select Open and you will see the Lakehouse explorer view of your lakehouse.

Explore lakehouse tables and schema

After discovering a data asset in the catalog, the next step is to explore its structure. Understanding the tables, columns, and data types helps you determine whether the data meets your needs.

  1. From the catalog, select your sales_data lakehouse to open it.

  2. In the Explorer pane on the left, expand the Tables folder to see the sales table you created earlier.

  3. Select the sales table. The main view shows a preview of the data, displaying the first several rows.

    Screenshot showing Delta Lake files for the sales table.

  4. In the table preview, observe the column headers: SalesOrderNumber, SalesOrderLineNumber, OrderDate, CustomerName, EmailAddress, Item, Quantity, UnitPrice, and TaxAmount. Below the preview, you can see the schema showing each column’s data type.

  5. Select File view to see the underlying files as well.

    Screenshot showing Delta Lake files for the sales table.

The table is stored as Parquet files in the Delta Lake format. The _delta_log subfolder contains transaction logs that track all changes to the table, enabling features like ACID transactions and time travel.

Create a shortcut to access data from another workspace

Shortcuts let you reference data from other workspaces without copying it, providing a live connection to the source. In this task, you create a second lakehouse and a shortcut pointing to the sales data in your original lakehouse.

  1. At the top of the page, select your workspace name to return to the workspace view.

  2. Select + New item > Lakehouse to create a second lakehouse. Name it analytics (or another name of your choice).

    This lakehouse represents your analytics workspace where you’ll transform data and build semantic models.

  3. When the new lakehouse opens, in the Explorer pane, select the … menu for the Tables folder and select New shortcut.

    Screenshot showing new shortcut option.

  4. In the New shortcut dialog, select OneLake as the shortcut source.

    Screenshot showing shortcut creation dialog with OneLake selected.

  5. In the workspace list, select the workspace containing your original sales_data lakehouse (for example, Data-Engineering).

  6. Select the sales_data lakehouse, then select the Tables folder.

  7. Select the sales table, and then select Next.

  8. Review the shortcut details before selecting Create.

    • Target type: Which shortcut type, OneLake in this case
    • Shortcut location: The lakehouse you want the shortcut in
    • Shortcut name: Defaults to sales
    • Source: The lakehouse where the data exists

    Screenshot showing shortcut confirmation with shortcut name and source location.

  9. In the Explorer pane of your analytics lakehouse, expand the Tables folder. You should see the sales table with a shortcut icon (a link overlay).

    Screenshot showing the sales table with a shortcut icon in the analytics lakehouse.

  10. Select the sales shortcut to view its data. The data is identical to the original table, but no data was copied. The shortcut provides a live reference to the source.

Query data using the SQL analytics endpoint

Every lakehouse includes a SQL analytics endpoint that provides read-only T-SQL access to tables. This endpoint enables you to query lakehouse data using familiar SQL syntax, making it easy to explore data and verify it contains what you need.

  1. At the top-right of your analytics lakehouse page, locate the drop-down menu that currently shows Lakehouse. Select the drop-down and switch to SQL analytics endpoint.

    Screenshot showing the switch from Lakehouse to SQL analytics endpoint.

    The view changes to a SQL-focused interface where you can query tables using T-SQL.

  2. In the toolbar, select New SQL query to open a query editor.

  3. In the query editor, enter the following T-SQL query:

     SELECT 
     Item,
     SUM(Quantity * UnitPrice) AS TotalRevenue,
     SUM(Quantity) AS TotalQuantity
 FROM sales
 GROUP BY Item
 ORDER BY TotalRevenue DESC;

    This query calculates total revenue and total quantity sold for each item.

  4. Select Run to execute the query. Review the results showing revenue by item.

    Screenshot showing SQL query results with revenue by item.

  5. Write a second query to find the top customers by revenue:

     SELECT TOP 5
     CustomerName,
     SUM(Quantity) AS TotalQuantity,
     SUM(Quantity * UnitPrice) AS TotalRevenue
 FROM sales
 GROUP BY CustomerName
 ORDER BY TotalRevenue DESC;

  6. Run the query and review the results.

Create and explore a semantic model

Semantic models provide a business-friendly layer on top of data, including predefined relationships, measures, and calculations for Power BI reports. In this task, you create a semantic model from the lakehouse and explore it using the Explore this data feature.

  1. While still in the SQL analytics endpoint view of your analytics lakehouse, select the New semantic model option in the toolbar.

    Screenshot showing the new semantic model option.

  2. In the Create a semantic model dialog, verify that the sales table is selected. Give the semantic model a name like Sales Analysis and select Create.

    After a moment, the semantic model is created. The model appears in your workspace.

  3. Navigate back to your workspace and find the Sales Analysis semantic model you just created. Select the ellipses (…) menu for the semantic model and choose Explore this data.

    Screenshot showing the Explore this data option for a semantic model.

    The Explore this data feature opens a lightweight exploration interface where you can create quick visualizations without opening Power BI Desktop.

  4. In the Explore window, drag the Item field to the canvas to create a table visualization.

  5. Drag the Quantity field to the Values area to show total quantities by item.

  6. In the Visualizations pane, change the visualization type to a Bar chart or Column chart to see a graphical representation.

    Screenshot showing the Explore window with a bar chart of quantities by item.

  7. Experiment with other fields and visualization types to explore the data.

(Optional) Use Copilot to write a SQL query

If you have access to Microsoft Copilot in Fabric, you can use it to assist with writing SQL queries. Copilot can generate queries based on natural language descriptions, helping you explore data faster.

Note: This task requires Copilot access in your Fabric environment. If Copilot is not available, skip this task.

  1. Return to your analytics lakehouse and switch to the SQL analytics endpoint view.

  2. Open a new SQL query editor.

  3. In the query editor, locate the Copilot button in the toolbar (it may appear as a lightbulb or Copilot icon).

  4. Select the Copilot button to open the Copilot pane.

  5. In the Copilot prompt, type a natural language request such as:

    Write a SQL query to show total revenue by month for the year 2026, sorted by month.

  6. Review the query that Copilot generates. It should look similar to:

     SELECT 
     FORMAT(OrderDate, 'yyyy-MM') AS Month,
     SUM(Quantity * UnitPrice) AS TotalRevenue
 FROM sales
 WHERE YEAR(OrderDate) = 2026
 GROUP BY FORMAT(OrderDate, 'yyyy-MM')
 ORDER BY Month;

  7. Select Insert to add the generated query to your editor, then select Run to execute it.

  8. Review the results. Try asking Copilot to write other queries, such as:

    • “Show me the average unit price by item”
    • “Find all orders with quantity greater than 10”

Clean up resources

In this exercise, you created a lakehouse, explored the OneLake catalog, created shortcuts for cross-workspace access, queried data using the SQL analytics endpoint, and created a semantic model.

Once you finished exploring your lakehouses and semantic models, you can delete the workspaces you created for this exercise.

  1. In the bar on the left, select the icon for your workspace to view all of the items it contains.
  2. In the toolbar, select Workspace settings.
  3. In the General section, select Remove this workspace.
  4. Repeat these steps for any additional workspaces you created (such as the analytics workspace).