When SOA was declared dead there was a spate of articles and blogs on why the architecture “died.” Most pundits came to the conclusion that like many innovations it wasn’t the technology to blame but rather people. Architects lacked the skills to properly leverage SOA; business stakeholders failed to look at SOA as a strategic architecture, choosing instead to use it as a tactical integration-solving solution; network and systems’ administrators did not understand the unique characteristics and issues a well-designed SOA raised within the network and on systems; and developers were loathe to “reuse” and “share” services despite alternate promises of carrots and sticks.

SOA was treated as primarily a development concern, and thus network and systems’ administrators blithely ignored it rather than embrace its architectural quirks. The combination of dumping responsibility for the change in the organization required for SOA to succeed on developer’s shoulders and the lack of concern from the rest of IT certainly led to if not its demise then its reign as King of the IT Hill.

And now comes Infrastructure 2.0 - dynamic infrastructure – and we’re doing it all over again.

While infrastructure 2.0 focuses primarily on, well, infrastructure from a network and application network perspective, it is not and should not be laid upon the shoulders of network and systems’ administrators. Developers and architects, too, are going to be necessary for organizations to really leverage dynamic infrastructure and architect a more efficient, on-demand data center.

It isn’t just infrastructure that needs to be more flexible, it’s the people who interact with that infrastructure. It’s all of IT – and that includes developers and architects. Infrastructure 2.0 is a lot like Web 2.0: it’s about collaboration and integration, and the sharing of information and resources across a wide variety of devices, systems, and applications. Web 2.0 doesn’t work if people aren’t involved, as it’s the people that drive the direction and utility of any given Web 2.0 application.

Infrastructure providers are offering up a plethora of features and functionality designed to enable a more agile IT. Not a more agile network, not a more agile application network, but a more agile IT. That’s because infrastructure 2.0 and emerging data center models leveraging virtualization like cloud computing require collaboration, feedback, and integration of all the people and roles in a data center. The automated, on-demand data center cannot be accomplished by any one group within IT. It has to be done collaboratively, and that means some amount of flexibility on the part of all those involved.

That includes developers. In fact, without developers this whole cloud-computing-data-center-of-the-next-generation can’t really come to fruition.

Developers and architects need to take a fresh look at infrastructure – from the network up to the application network – and reconsider how innovations in both can be better leveraged within an architecture or even just within a single application. Also necessary is to view the application infrastructure as a partner in the application deployment process, one that can add significant value to the application in terms of security, performance, and availability.

Infrastructure 2.0 is specifically being designed to integrate with the rest of the application and network ecosystem via standards-based APIs. Some are service-enabled (SOA) and others are REST-based, but all take advantage of technology that has traditionally been under the domain of expertise of the developer. These APIs are of the same style – and intent – of those found in the wild on the “tubes” being used by Twitter, Facebook, and others to provide integration and management with other Web 2.0 applications.

But like a dynamic infrastructure’s API, if developers don’t look at them with eye toward how they can be leveraged, they’re useless. The  “agility” of dynamic infrastructure thus lies dormant and fails to bring about the change in architecture and the organization necessary for success. Just like SOA.

Maybe at first its just realizing the impact of the infrastructure on application scalability; on how the deployment of an application into an on-demand environment can affects its execution. Maybe it’s actually realizing that there are ways that the application, which is the crowning glory of the data center, can drive the on-demand nature of the entire data center architecture and collaborate with the infrastructure to ensure applications are as fast and secure and integrated as possible.

Maybe it’s figuring out how to communicate the unique information applications and their developers hold with the rest of the infrastructure such that the dynamic nature of infrastructure 2.0 can be leveraged to seamlessly handle the rapid rate of change that’s occurring – and that will continue to occur – within the data center and its applications.

We talk a lot about the agility and flexibility of Infrastructure 2.0 solutions; of their context-awareness and intelligence, of their ability to collaborate with the rest of the infrastructure. That infrastructure includes applications and their containers, their platforms, and must cast a wide enough net to include developers in the design of the next generation architecture if dynamic data centers are to come to fruition.

Developers need to understand how emerging data center models – and application models – are changing the very nature of the data center and their responsibilities. It’s no longer enough to just trust that a network and an application network will “be there” to deliver the applications developed. It’s no longer enough to toss applications over the wall onto a network and rely upon the network and systems’ administrators to sort through the morass when issues arise.

Everyone in IT needs to be more agile, more flexible, and understand the capabilities of the infrastructure to communicate and collaborate with one another, and then collaborate themselves on ways to leverage that dynamism in a way that makes it possible to keep innovating and addressing changes as they occur.

Without agile people, agile infrastructure is no more a benefit to the data center than traditional infrastructure. Like any tool, it must be used in order for it to provide value, and the only way to use those tools is to understand what their capabilities are and include them as a possible solution/platform/option when architecting and developing applications.

Related blogs & articles:

• 9 Ways to Use Network-Side Scripting to Architect Faster, Scalable, More Secure Applications
• Are admins developers too?
• How to instrument your Java EE applications for a virtualized environment
• Twitter’s API Limit: Static control in a dynamic world
• The Death of SOA Has Been Greatly Exaggerated
• SOA Isn’t Dead But Its Standards Are
• SOA is Dead. Long Live Services.
• I am wondering why not all websites enabling this great feature GZIP
• Cloud Computing: Is Your Cloud Sticky? It Should Be