Deciphering Data Architectures

Chapter 4. The Relational Data Warehouse

By the mid-2000s, I had used relational databases for years, but I had never been exposed to relational data warehouses. I was working as a database administrator (DBA) for a company that used an accounting software package to manage its financial transactions. The reporting from the package was limited and slow, so the company was looking to improve performance, create dashboards to slice and dice the data, and combine its financial data with data from a homegrown application to get a better understanding of the business.

My employer hired a consulting company to build this thing called a “relational data warehouse”—and, in a decision that changed the course of my career, asked me to help. We generated dashboards that saved the end users a ton of time and added business insights they’d never had before. When I saw the excitement on their faces, I knew I’d found my new passion. I changed my career to focus on data warehousing and never looked back.

What Is a Relational Data Warehouse?

A relational data warehouse (RDW) is where you centrally store and manage large volumes of structured data copied from multiple data sources to be used for historical and trend analysis reporting so that your company can make better business decisions. It is called relational because it is based on the relational model, a widely used approach to data representation and organization for databases. In the relational model, data is organized into tables (also ...