Big data is certainly one of the biggest buzz phrases in IT today. Combined with virtualization and cloud computing, big data is a technological capability that will force data centers to significantly transform and evolve within the next five years. Similar to virtualization, big data infrastructure is unique and can create an architectural upheaval in the way systems, storage, and software infrastructure are connected and managed. Unlike previous business analytics solutions, the real-time capability of new big data solutions can provide mission critical business intelligence that can change the shape and speed of enterprise decision making forever. Hence, the way in which IT infrastructure is connected and distributed warrants a fresh and critical analysis.

This book provides a general overview of traditional Hadoop architectures designed to deliver high-performance and scalable big data analytics. It is intended to provide a basis of understanding for interested data center architects and as a starting point for a deeper implementation engagement. This document assumes little to no background in big data or horizontally scaled query infrastructure, but rather it represents a starting point for the big data journey. Links to additional resources are provided at the end of this paper that help provide logical steps in this journey. Ultimately, the goal is to empower IT to enable an infrastructure that can provide immediate and deep business intelligence for decision making and agility. Every journey begins with a single step, and this document is the first step in recognizing the value of the big data process.

with careful planning and predetermined expectations, creating an optimized big data deployment is relatively straightforward. keep in mind only three or four years ago, broad commercial appeal for big data implementations was not a key requirement in data center design. However, you should be developing this infrastructure with an eye toward the horizon. Adopting technologies and underlying infrastructure, including networking that can provide the scale, performance, and headroom for tomorrow’s technologies is critical in offering the highest levels of investment protection, business agility, and time to market.