Re-Defining Enterprise Architecture for a New Era

Share it on Twitter  
Share it on Facebook  
Share it on Linked in  
Slide Show

Big Data Disruptions Can be Tamed with Enterprise Architecture

It seems that in many enterprise circles the terms "infrastructure" and "architecture" are interchangeable. It's an easy mistake considering both terms often refer to everything from physical hardware, multiple software layers and virtual/cloud environments.

However, a simple way to look at it is infrastructure is what you have while architecture is how you use it.

Tech consultant Kyle Gabhart hits upon this crucial difference by pointing out that you can't actually deploy an enterprise architecture. On a basic level, you can only deploy a technical architecture, which is to say infrastructure, and from that create both a solution architecture and an enterprise architecture as abstracted layers. All three are crucial components of a well-ordered data environment because without effective EA and SA, changes to the infrastructure layer can only occur in isolation and will ultimately fail to meet the requirements of the organization.

According to Gartner, a good enterprise architecture will feature five crucial "deliverables" in order to provide optimal business value. These include measurable ones that gauge the impact of EA on business, actionable ones that drive change and are closely tied to outcomes and requirements, as well as diagnostics, enablers and operational deliverables overseeing change management, information management and overall functionality. Once these elements are in place, the challenge is to find the proper balance of time, energy, capital and other resources while recognizing that most people outside of IT care only about measurable and actionable deliverables.

Of course, the need for enterprise architecture stems from the fact that enterprise infrastructure is growing more complex, even without factoring in virtualization and the cloud. As GigaOm's James Urquhart points out, the notion of establishing a single blueprint for modern application systems is impossible now that individual users are able to spin up new resources with little or no regard for broader organizational needs. This means the enterprise architect has to shift focus from maintaining stable computing models to fostering greater alignment between applications, services and data sets. Think of it as the end of the enterprise architect per se and the beginning of the enterprise product manager.

Regardless of the shifts in technologies or personnel responsibilities, however, there is one key system that will remain an integral component of enterprise architecture: the governance framework. As ebizq's Adrian Grigoriu notes, governance oversees not only the systems and technologies in place, but the rules regarding use, the security procedures and platforms that keep them safe and a broad range of other functions. In short, governance determines what decisions are made regarding the enterprise environment, who makes them and on what criteria.

Enterprise architecture, and the people who oversee them, are therefore crucial components in the drive to shed the old silo environments that have arisen in most enterprises in favor of the flatter, more federated and highly dynamic systems of the future.

Just as a building is more than a collection of concrete, steel and bricks, the enterprise is more than just servers, storage and networks.