Data Management at Scale

Chapter 6. Connecting the Dots

This chapter will quickly recap the architectures, using different viewpoints, then connect them to the data management disciplines of governance, data modeling, and metadata management. These are interlinked, and we want to ensure that security, governance, metadata, and data modeling are consistently and uniformly applied on all architectures.

From there, the chapter will discuss data and interface design for all architectures. You will learn when to choose one integration pattern or another, what combinations you can make, and what works best in hybrid and multicloud models. I will also discuss important standards for discoverability and interoperability and the benefits of setting principles for stable and reusable data, which improve overall data consumption and remove repeatable work from the domains. Additionally, you will learn why such tremendous effort goes into making the architecture metadata-driven. Finally, I’ll look at how to strive for semantical consistency by connecting your domain endpoints to an abstraction layer that describes each of the unique datasets and elements.

Recap of the Architectures

In Chapter 2, we started with the holistic picture, including all communication flows. You learned that internal domain and application complexity must be hidden from other domains and that consistent consumption-optimized data must be exposed via the data layer.

All applications that need to communicate and exchange data are brought ...