Data Lakes For Dummies

Introduction

In December 1995, I wrote an article for Database Programming & Design magazine entitled “I Want a Data Warehouse, So What Is It Again?” A few months later, I began writing Data Warehousing For Dummies (Wiley), building on the article’s content to help readers make sense of first-generation data warehousing.

Fast-forward a quarter of a century, and I could very easily write an article entitled “I Want a Data Lake, So What Is It Again?” This time, I’m cutting right to the chase with Data Lakes For Dummies. To quote a famous former baseball player named Yogi Berra, it’s déjà vu all over again!

Nearly every large and upper-midsize company and governmental agency is building a data lake or at least has an initiative on the drawing board. That’s the good news.

The not-so-good news, though, is that you’ll find a disturbing lack of agreement about data lake architecture, best practices for data lake development, data lake internal data flows, even what a data lake actually is! In fact, many first-generation data lakes have fallen short of original expectations and need to be rearchitected and rebuilt.

As with data warehousing in the mid-’90s, the data lake concept today is still a relatively new one. Consequently, almost everything about data lakes — from its very definition to alternatives for integration with or migration from existing data warehouses — is still very much a moving target. Software product vendors, cloud service providers, consulting firms, industry analysts, ...