Overview

After configuring Unity Catalog, the next step is to connect your Raw Data Landing Container in Azure Data Lake Storage Gen2 to Unity Catalog.

This guide demonstrates how to:

• Create an External Location for the raw data container
• Create a Raw Schema
• Create an External Volume
• Prepare the Bronze (Raw) layer for data ingestion

The External Volume enables Unity Catalog to securely access raw datasets stored in Azure Data Lake Storage Gen2 using Managed Identity.

Learning Objectives

After completing this guide, you will be able to:

• Connect a Raw Data container with Unity Catalog
• Create an External Location
• Create a Raw Schema
• Create an External Volume
• Store structured, semi-structured, and unstructured files securely
• Prepare the Bronze Layer for ETL pipelines

Architecture

Raw Landing Zone

The ecomm-raw-data container acts as the landing zone for all source system data.

Files are stored in their original format before any transformation.

Typical files include:

• CSV
• JSON
• Parquet
• XML
• TXT
• LOG
• Excel
• Images
• Other binary files

Example structure:

ecomm-raw-data
│
├── customers/
│      ├── customers.csv
│
├── orders/
│      ├── orders.csv
│
├── products/
│      ├── products.json
│
├── inventory/
│      ├── inventory.parquet
│
├── application_logs/
│      ├── app.log
│
└── images/
       ├── product001.jpg

Step 1 — Create an External Location

Navigate to:

Catalog
   ↓
External Locations
   ↓
Create External Location

Configure the External Location using the following values.

After entering all the required information, click Create.

Why create an External Location?

An External Location securely maps the ecomm-raw-data container to Unity Catalog.

Instead of using Storage Account Keys, Unity Catalog authenticates using the configured Managed Identity through the Storage Credential.

This enables secure, governed access to raw files for notebooks, jobs, and ETL pipelines.

Step 2 — Create Raw Schema and External Volume

Open a SQL Notebook and execute the following commands.

%sql
USE CATALOG ecommerce; 
CREATE SCHEMA IF NOT EXISTS raw; 
CREATE EXTERNAL VOLUME IF NOT EXISTS raw.raw_landing 
LOCATION 'abfss://ecomm-raw-data@stecommercedeveastus001.dfs.core.windows.net/' COMMENT 'Landing zone for raw data'; 

SQL Explanation

Select the Catalog

USE CATALOG ecommerce;

Sets ecommerce as the active Unity Catalog.

Create the Raw Schema

CREATE SCHEMA IF NOT EXISTS raw;

Creates the raw schema if it does not already exist.

The raw schema represents the Bronze Layer of the Medallion Architecture.

Create the External Volume

CREATE EXTERNAL VOLUME IF NOT EXISTS raw.raw_landing
LOCATION 'abfss://ecomm-raw-data@stgpunithadlsdevus001.dfs.core.windows.net/'
COMMENT 'Landing zone for raw data';

Creates an External Volume named raw_landing inside the raw schema.

The volume points directly to the ecomm-raw-data container in Azure Data Lake Storage Gen2.

This location stores incoming files exactly as they arrive from source systems.

After execution, the notebook should display:

OK

This confirms that the schema and external volume have been created successfully.

Databricks Notebook

This guide is implemented using the following Databricks notebook.

Notebook Overview

The notebook performs the following tasks:

• Select Unity Catalog
• Create the Raw Schema
• Create the External Volume
• Configure the Bronze Landing Zone
• Validate successful execution

Final Resource Hierarchy

Azure Subscription
│
├── Azure Databricks Workspace
│
├── Unity Catalog
│      │
│      └── ecommerce
│              │
│              └── raw
│                      │
│                      └── raw_landing
│
├── External Location
│      │
│      └── exloc-ecomm-raw-eastus
│
└── Azure Storage Account
        │
        └── ecomm-raw-data
                │
                ├── CSV
                ├── JSON
                ├── LOG
                ├── XML
                ├── Parquet
                ├── Excel
                └── Images

Data Flow

Source Systems
      │
      ▼
CSV / JSON / XML / LOG / Images
      │
      ▼
Azure Data Lake Storage Gen2
(Container : ecomm-raw-data)
      │
      ▼
External Location
      │
      ▼
Unity Catalog
      │
      ▼
Schema : raw
      │
      ▼
External Volume : raw_landing
      │
      ▼
Bronze Layer

Verification Checklist

Best Practices

• Use a dedicated container for raw data.
• Store source files in their original format.
• Do not modify raw files.
• Organize files by source system.
• Use Managed Identity instead of Storage Account Keys.
• Apply Unity Catalog permissions to control access.
• Maintain separate Bronze, Silver, and Gold layers.
• Use descriptive names for schemas and volumes.