To help you get started, here’s a sample process to follow.
Note: Average response time for report loading is 10-15 seconds.
Execute a guided validation process across Fabric Workspace, Lakehouses, Notebooks, Power BI, Databricks, and Purview to ensure seamless data movement from Bronze to Gold layers, validated reporting, and complete data lineage visibility.
- Task: Validate 4 folders are created successfully in the Fabric Workspace
- databricks
- lakehouses
- notebooks
- reports
Note: The databricks folder is included only when the user has completed deployment and configuration of Azure Databricks for Fabric integration.
- Task: Browse the resources in each folder and subfolders.
- Task: Open notebooks folder and verify below resources:
- bronze_to_silver: Contains 16 notebooks to process data in bronze lakehouse and populate data models in the silver lakehouse
- data_management: Contains 5 notebooks to truncate and delete tables if a process needs to be rerun
- schema: Contains 8 notebooks to create data models
- silver_to_gold: Contains 19 notebooks to take data from silver tier with necessary processing and insert data to tables in gold lakehouse
- Observe: You will also see two runner notebooks:
- run_bronze_to_silver: This notebook runs all the notebooks related to data processing from bronze to silver
- run_silver_to_gold: This notebook runs all the notebooks related to data processing from silver to gold
- Task: Navigate to the notebooks/schema folder in the Fabric workspace.
- Task: Verify that notebooks are available to create data models for the following domains:
- Customer
- Product
- Sales
- Finance
- Task: Open the targeted notebook 'model_shared_gold' in the Fabric workspace.
- Task: Confirm that the notebook interface loads successfully without errors.
- Task: Review the notebook content to verify the database schema and table structures referenced or defined within it.
- Task: Navigate to the lakehouses folder and verify that three lakehouses are displayed.
- Task: Open the udplz_gold lakehouse and confirm the schema and list of tables are visible.
- Task: Access the SQL endpoint associated with udplz_gold.
- Task: Run T-SQL queries (e.g., SELECT COUNT(*) FROM [udplz_gold].[salesadb].[order]) and verify that valid row counts are returned.
- Task: Navigate to the notebooks folder.
- Task: Open the bronze_to_silver subfolder.
- Observe: list of related notebooks.
- Observe: Open any notebook verify that the notebook details are loaded correctly.
- Task: Navigate to the notebooks folder.
- Task: Opens the silver_to_gold subfolder.
- Observe: list of related notebooks.
- Task: Open any notebook verify that the notebook details are loaded correctly.
- Task: Navigate to the notebooks folder.
- Task: Open the data_management subfolder.
- Observe: list of related notebooks.
- Task: Open any notebook verify that the notebook details are loaded correctly (e.g., truncate_all_tables_gold)
- Task: Navigate to the notebooks folder.
- Task: Verify that two runner notebooks is displayed.
- Task: Open any notebook and verify that the notebook details are loaded correctly.
- Task: Open the Fabric workspace and open reports folder and navigate to sales_dashboard.
- Observe: YOY Net sales Comparison chart.
- Observe: trend line and identify any downward trend in the data.
- Task: Hover over a specific year on the trend line(e.g. 2022)
- Observe: Verify that the exact Net Sales value for that year is displayed in the tooltip.
- Observe: Revenue Distribution by Customer Segment visual.
- Observe: Review the distribution across customer segments:
- Individual
- Business
- Government
- Task: Compare the thresholds and contribution of each segment to overall revenue.
- Task: Open fabric workspace and navigate to databricks folder
- Task: Navigate to udplzcatalog mirrored Azure Databricks and validate the list of tables under sales schema - order, orderline and orderpayment
- Task: Open Purview and select Data Map
- Task: Navigate Data Sources and validate Fabric Data source is Available in Collection created
- Verify Fabric is successfully connected to Microsoft Purview, and Purview collections appear in Fabric.
Scan and Report
- Task: Navigate to Data Map -> Monitoring
- Observe: Scan details and see information on Cards and Graph
Data Lineage and metadata review
- Task: Navigate to Unified Catalog -> Catalog management -> Data Products
- Task: Select Data Product Sales Resources.
- Task: Open udplz_gold. Observe Data asset lineage
Unified Data Platform using Microsoft Fabric, Purview, and Azure Databricks for integrated, governed, and analytics-ready insights.
