Databricks Data Warehouse Project: Star Schema & OLAP Analysis

Overview

This project demonstrates Data Warehouse design and OLAP analysis using PySpark in Databricks.
It follows a Star Schema approach with one fact table (sales_fact) and four dimension tables:

• date_dim
• product_dim
• customer_dim
• store_dim

The workflow includes:

1. Synthetic Data Generation with Faker
2. Data Modeling in a Star Schema format
3. Delta Table storage to simulate Data Warehouse persistence
4. OLAP-style queries for business insights
5. ER Diagram visualization of the schema

---

Technologies Used

• Databricks (PySpark runtime)
• Delta Lake for table storage
• Faker for synthetic data generation
• Matplotlib & NetworkX for ER diagram visualization
• Data Modeling with Star Schema

---

Example Queries

# Total Sales by Category
spark.table("sales_fact") \
    .join(spark.table("product_dim"), "ProductID") \
    .groupBy("Category") \
    .agg(sum("Sales_Amount").alias("Total_Sales")) \
    .orderBy(desc("Total_Sales")) \
    .show()