Tutorial: Introduction to Graph in Microsoft Fabric

Note

This feature is currently in public preview. This preview is provided without a service-level agreement, and isn't recommended for production workloads. Certain features might not be supported or might have constrained capabilities. For more information, see Supplemental Terms of Use for Microsoft Azure Previews.

This tutorial is a step-by-step walkthrough of an end-to-end graph scenario in Microsoft Fabric. It covers everything from creating your graph model to querying your data and analyzing insights. Complete this tutorial to build a basic understanding of the Microsoft Fabric graph experience and its capabilities.

Prerequisites

Before you start this tutorial, verify that:

1. Graph is available in your region.

2. Graph is enabled in your Fabric tenant.

   

3. You're a member of a Fabric workspace or have permission to create items in the workspace. For more information, see Workspaces in Microsoft Fabric.

   Important

   Access management of the graph is restricted to the workspace that hosts it. Users outside of the workspace can't access the graph. Users within the workspace who have access to the underlying data in the lakehouse can model and query the graph.

Graph end-to-end scenario

In this tutorial, you take on the role of a data analyst at the fictional Adventure Works company. You build a graph model to represent the relationships between customers, orders, employees, products, and vendors. Then, you query the graph to uncover insights about customer purchasing behavior and product performance. Follow these steps:

1. Load sample data into a lakehouse.
2. Create a graph by creating a graph model and loading data from OneLake.
3. Add nodes to your graph for each entity in the data model.
4. Add edges to your graph to define relationships between nodes.
5. Query the graph with the query builder using an interactive visual interface.
6. Query the graph with GQL using the code editor.
7. Clean up tutorial resources by deleting the workspace and other items.

Graph end-to-end architecture

• Data sources - Microsoft Fabric makes it quick and easy to connect to Azure Data Services, other cloud platforms, and on-premises data sources.

• Storage - Microsoft Fabric standardizes on Delta Lake format stored in OneLake. Graph in Microsoft Fabric reads data directly from your lakehouse tables to construct the graph model.

• Graph modeling - Create nodes and edges that represent entities and relationships in your data. The graph model provides a flexible way to explore connected data.

• Query and analyze - Use the visual query builder for interactive exploration or write GQL (Graph Query Language) queries for more complex analysis.

Sample data

For this tutorial's sample data, use the Adventure Works sample dataset. Adventure Works is a fictional bicycle manufacturer that sells bicycles and accessories to customers worldwide.

Note

The Adventure Works dataset used in this tutorial is a custom-transformed version designed specifically for demonstrating Graph capabilities. It differs from standard Adventure Works datasets and supports graph-specific features.

The Adventure Works dataset includes:

• Customers - People who purchase products
• Orders - Sales transactions
• Employees - Staff who process sales
• Products - Items available for purchase
• Product categories and subcategories - Product classification hierarchy
• Vendors - Suppliers who produce products

Data model

The Adventure Works data model demonstrates a typical retail scenario with multiple entities and relationships. In this tutorial, you model the following relationships:

Relationship	Description
Employee sells Order	Employees process customer orders
Customer purchases Order	Customers make purchases
Order contains Product	Orders include products
Product isOfType ProductSubcategory	Products belong to subcategories
ProductSubcategory belongsTo ProductCategory	Subcategories belong to categories
Vendor produces Product	Vendors supply products

Next step

Load sample data