Managing Cloud Native Data on Kubernetes

Chapter 9. Data Analytics on Kubernetes

Progress in technology is when we have the ability to be more lazy.

Dr. Laurian Chirica

In the early 2000s, Google captivated the internet with a declared public goal: “to organize the world’s information and make it universally accessible and useful.” This was an ambitious goal and accomplishing it would, to paraphrase, take “computer sciencing” the bits out of it. Given the increasing rate of data creation, Google needed to invent (and reinvent) ways of managing data volumes no one had ever considered. An entirely new community, culture, and industry were born around analyzing data called analytics, tackling what was eventually labeled “big data.” Today, analytics is a full-fledged member of almost every application stack and not just relegated to a Google problem. Now it’s everyone’s problem; instead of an art form restricted to a small club of experts, we all need to know how to make analytics work. Organizations need reliable and fast ways to deploy applications with analytics so that they can do more with less.

The laziness Dr. Chirica was talking about in a tongue-in-cheek way in the quote that opens this chapter describes an ideal future. Instead of having a hundred-person team working night and day to analyze a petabyte of data, what if you could reduce that to one person and a few minutes? The cloud native way of running data infrastructure is a path we should all work toward to achieve that kind of glorious laziness.

We’ve already ...