Data Management at Scale

Chapter 8. Turning Data into Value

In the previous chapters you learned what it takes to make data available in a safe and controlled way with governance and security. In this chapter you will learn how to turn data into value. This is the most complicated part.

Business requirements always come first. Turning data into insights or actions requires understanding how information flows and using that value to identify business opportunities. Use cases may start as one-off projects, but ideally you’ll turn your key data into maintainable solutions that deliver constant value for the organization. Depending on your business requirements, you might use different techniques, such as business intelligence, real-time decision making, or machine learning.

Then there are nonfunctional requirements. It takes many different database technologies to accommodate a large variety of complex use cases. Depending on the use case, you might pick one or several of these. Additionally, there are variations in the types of transformations, their velocities, optimizations for parallelization, and consumption patterns, all of which affect your end result. Finally, performance limitations could force you to duplicate or restructure data in favor of reading it faster.

The suggested approach for all of these challenges is to create standardized, reusable patterns and building blocks. These will be built upon the data distribution foundation, discussed in Chapter 6, and work with the governance model, discussed ...