Unlocking the Value of Real-Time Analytics

Chapter 3. Key Components of the Real-Time Stack

Chapters 1 and 2 looked at requirements of real-time analytics and examined some of the ways companies attempt to optimize their existing infrastructure to increase concurrency and lower the latency from user query to response. However, we haven’t examined why a real-time stack is so vital to business. We also need to understand what components are needed to achieve these requirements. In this chapter, we will move from the current state of analytics to understanding the reasons for real-time analysis and identifying the tools needed to reach a faster, more dynamic level.

Driving Forces of Real Time

It is easy to understand why companies and organizations strive to report as quickly as possible on the transactional data they generate. The closer they are to when changes happen, the faster they can adapt to the change. More importantly, real time is a shift in the thought process of business. How can your business shift through these stages:

Historical (descriptive) reporting

What happened and when?

Diagnostic analytics

Why did these things happen?

Predictive analytics

When will this thing happen again?

Preemptive analytics

How do we stop this thing from happening again?

There are many examples, especially in recent years, that illustrate this need.

The Internet of Things (IoT) demanded a dramatic shift in how companies and organizations capture data. Everyday objects are starting to become ingrained with data-collecting ...