In the last 11 chapters, we have been through a journey exploring the data ecosystem, trying to define a reference architecture for a data intensive system that consisted of lots of micro services working in a decoupled manner and interacting with each other using an event-driven mechanism and a data pipeline. We looked at various data processing use cases of batch processing and stream processing. We also discussed data storage solutions for huge datasets. But if I tell you that all this will fail if you don't get your dissemination architecture right, you will probably frown at me. It's a hard-to-digest fact.

In this chapter, we will look at the following topics:

• Dissemination architecture and what problem it tries to solve
• Planning a dissemination architecture, including what you should keep in mind when defining a dissemination architecture
• Various technical and architectural approaches that enable an efficient and resilient data-dissemination architecture
• Technologies that enable efficient data-dissemination
• Putting data dissemination into the bigger overall picture

Before delving into detail around the dissemination architecture, let's refresh our memory a little. In Chapter 2, Defining a Reference Architecture for Data-Intensive Systems, we had defined a component architecture for a data intensive architecture that looked like this:

When we talk about data dissemination, we really are talking about data access and data insight components of this architecture. The data access components' primary objective is to provide an efficient means for the data to reach its intended audience. Data insight refers to the ability to make sense out of the data that has been disseminated to the end user.

Unless mentioned otherwise, data access and data insight go hand in hand when defining a dissemination architecture.

But what does the dissemination of data actually mean?