Automating the Modern Data Warehouse

Introduction

For the enterprise, cloud is two things: first, it’s a force for potential business and IT transformation; second, it’s a mechanism for reducing or controlling costs. Even though enterprises have tended to emphasize the latter at the expense of the former, the most compelling reason to migrate an on-premises data warehouse to the cloud has little to do with reducing costs or taking advantage of cloud’s tax-friendly OpEx. In fact, a focus on cost misses what is new and transformative about cloud.

A better, more compelling rationale for cloud migration is to modernize and improve the data warehouse by substantively automating it.¹ For one thing, automation of this type—and at this scale—has the potential to eliminate time-consuming, tedious, and rote tasks. For another, it frees up bright, imaginative, creative human technicians—DBAs, ETL developers, business intelligence (BI) developers, architects, etc.—to focus on more challenging problems.² Another benefit of automation at this scale is that it permits IT and the lines of business to respond more rapidly to changing conditions: it becomes possible and cost-effective to accommodate one-off, seasonal, or unprecedented workloads and use cases.

This gets at the best rationale for migrating—one that is grounded in the logic of data warehouse modernization itself: to transform the business. A focus solely on cost savings sets an organization up to make poor choices in the future. A focus on transforming the business by ...