Networking’s Virtualization Opportunity

Cloud computing and virtualization, undoubtedly two of the most overused industry buzz words, both are finally showing signs of real market momentum.

Service providers like AT&T with its Synaptic Hosting service and content providers like Amazon with its Elastic Compute Cloud (EC2) environment are now deploying new cloud computing services. Many businesses are trading standardized IT applications for converged ones, and in some cases the “cloud” is replacing aspects of the traditional IT team. If you consider that about 80 percent of an enterprise’s local computing resources go unused, the virtualization of certain IT applications certainly makes sense.

Consequently, I keep asking myself: What impact will these trends have on the network, and what new requirements will they demand?

Applications are rapidly becoming services; in other words, something that end users are willing to pay for. Software as a Service (SaaS) is a perfect example. I use a SaaS application, Salesforce.com, nearly everyday. And guess what? It doesn’t reside anywhere inside the walls of my company – it’s “in the cloud.” I access it through a service provider’s network.

The service provider I use to access my cloud-based applications varies by my location – home, office, or on the road. Does the quality of my experience vary from one “application access point” to another? Of course it does. However, that quality has little to do with the application provider and everything to do with the service provider. As a result, service providers are inherently becoming the “application access provider” and that’s not necessarily a good thing unless they generate incremental revenue from that function.

Is this something that service providers should be concerned about? Absolutely. There’s a huge shift occurring in the “perceived value” that end users place on the service provider and the application provider. And right now that perceived value is with the latter.

Virtualization, and the cloud computing capabilities it enables, impacts every segment of a service provider’s business. And, both trends offer compelling benefits for end users, like reducing time to market, lowering barriers to entry in new markets and providing greater flexibility to scale business processes seamlessly and cost effectively.

And what about mobility? Increasing ease of connectivity to the mobile Web and subsequent network availability expands the reach of virtualized services beyond just the wireline connection. Take the iPhone for example. When Apple released the iPhone, Google reportedly experienced a 50-fold increase in mobile-originated search traffic – due to the ease of use, improved experience and third-party, open application innovation inherent in the new device. As others release similar devices, like T-Mobile’s G1 and the first phone to use Google’s Android open source mobile platform – and as 3G/4G bandwidth increases – this will have a huge impact on service providers’ business models and their networks.

So, what does all this mean for the service provider? Put simply, they need to change the way they compete by adapting their traditional network build-out and service delivery model to address this new environment.

For starters, the network must be optimized for the time-to-market battle – with increased service awareness in the access and metro for accelerated service creation, activation and assurance to obtain a competitive advantage more quickly.

They must also be able to differentiate services with SLAs for various application requirements, which will enable them to capture more revenue per customer. To do that, they need to provide more deterministic service behavior with capabilities like hierarchical QoS and improved service stratification.

And, because no one knows what the next big cloud-based bandwidth driver will be, service providers need to be able to respond to unpredictable demand curves – or one-time, transactional demands – with the ability to dynamically allocate, increase, and/or decrease network capacity with low touch via programmability.

At the same time, carriers must also consider optimizing their networks across the wireline and wireless portions to facilitate ubiquitous connectivity and seamless service delivery.

And finally, in this virtualized world, service providers need to ensure that all network elements are working in harmony and at their peak performance potential. That means they need integrated and advanced service-aware data, control and management plane capabilities from the customer edge all the way to the service provider core.

One thing is clear: Service providers’ current approach to network building and service delivery won’t be able to withstand these emerging application innovation models in the long run. Providers need a fundamentally different perspective and approach to the network that replaces today’s ”dumb pipe” application access connectivity with intelligent, automated and service-driven networks to help them monetize and meet end-user needs in this new application-driven era.

Dave Parks is director of product marketing at Ciena Corp. (CIEN). Ciena provides flexible platforms, intelligent software and professional services to help service providers and enterprises worldwide use their networks to change the way they compete.

Share this article: Email, Slashdot, Digg, Del.icio.us, Yahoo!MyWeb, Windows Live Favorites, Furl
Add this article feed to: RSS, My Yahoo, Newsgator, Bloglines