In this section we look at the direct evolution of the original ideals of distributed storage and MapReduce processes, and how Spark builds on them into something more flexible, accessible and mature.

We'll introduce some concepts and outline how Spark fits into several workloads, checking back against our data platform wishlist as we go.