We will now see how modern data architecture addresses the pain points of a legacy EDW and prepares the organization to handle the big wave of data. It is designed to handle both structured and unstructured data in a cost effective and scalable mode. This provides the business with a wide range of new capabilities and opportunities to gain insights.

Instead of a complete EDW replacement, this architecture leverages the existing investment by preserving end-user interfaces that require relational stores. In this model, Hadoop becomes the prime data store and EDW is used to store aggregates.

The following figure shows you how to transition from legacy EDW-based solutions to a hybrid Hadoop-based ecosystem, where EDW's role is reduced to hosting the aggregated data enabling queries via well-established tools that are relational in nature:

The following figure shows you the new reference architecture for a Hadoop-based Data Lake:

Let...