After a steady stream of significant investments in both labor and dollars as well as a series of successful trials, network function virtualization has proven its value to service providers. NFV brings virtualization in service providers’ back-end data centers, effectively converging IT and network operations in order to adopt modern concepts like development operations. NFV can reduce service providers’ costs and streamline operations and directly lead to better, higher-value services and faster time to market for customers.
The next step in the rapid evolution of NFV is expansion outside the realm of the service provider to the enterprise premises. NFV is getting ready for deployment into the enterprise network as an enabler of added-value services beyond increasingly commoditized network connectivity. The perceived challenge is that selling these value-added services to enterprises will be difficult because CIOs and network administrators will be reluctant to hand more control over to the carrier. However, many compelling reasons exist to convince them to take that leap.
The term NFV emerged from the network service provider industry. NFV enables service providers to embrace virtualization. In their back-end data centers, NFV effectively enables the convergence of IT and network operations in order to adopt modern concepts like devops. In the context of enterprise network services, NFV is known for the so-called virtual customer premise equipment use-case.
One primary reason NFV is getting ready for implementation in the enterprise setting is that compute power has caught up with the capabilities that NFV provides to carriers: the ability to completely automate the delivery of services. The elimination of the human element also means the elimination of human error that leads to outages, which enables cloud-grade instantaneity, flexibility and elasticity in network services. VMware’s pioneering efforts to virtualize the data center proves you can share resources and still have your own sandbox. You reap the benefits of doing away with on-site maintenance requirements and of the resulting cost reductions without having to sacrifice performance. Network services based on NFV will not only be more flexible and elastic, they will be more reliable, less expensive and faster to provision to customers.
NFV can enable providers to deliver the various and disparate network services that have traditionally been delivered as hardware appliances, firewalls, load-balancers and application delivery controllers, wide-area-network optimization, etc. NFV enables them to be offered as virtualized, software-based services running on standard hardware. Just as with virtualizing servers, virtualizing network functions offers a host of flexibility, elasticity and instantaneity benefits.
Another fundamental benefit is the enablement of service-chain automation – critical to lower operations and change-management costs and eliminates the risk of human error. This automation is the real game changer, and NFV and programmability are the critical means to achieving it.
Virtualized network functions allow organizations to dynamically provision application aware networks wherever they are needed, on an on-demand basis. Just as virtualization has completely revolutionized the data center, NFV is poised to do the same for the network.
The carrier that delivers value-added services via NFV will not only gain a competitive advantage over other carriers, but also over the so-called “eyeball” network providers, like Comcast and Cox, that now offer businesses – and consumers – high-speed Internet and voice-over Internet Protocol connections bundled with security software and other solutions from third-party vendors.
Business class vs. coach
A good analogy is the way in which airlines make money. Winning the lowest fare war is impossible; the airlines large and small just keep undercutting each other on price for those middle-coach seats. They have very low margins and that will never change.
Where airlines make money is by selling premium business-class services. More leg room, free movies and drinks, and better food – all value-added services customers are willing to pay for. Even if a customer does not want everything and only cares about more leg room, the customer pays the full business-class price. The customer can’t opt out of the free movie and pay less; the fee and delivery are controlled by the airline.
This same model applies to carriers. All carriers offer the connection circuitry; so without any real differentiating factors, the margins are very low. Value-added services, such as the ability to offer a “secure network,” “guaranteed service level agreements” and “complete network visibility” are the moneymakers.
These value-add services are best offered via NFV as it provides the carriers with the ability to perfectly align the network services with the business needs of enterprises. NFV flexibility, elasticity and instantaneity mean a network with an unlimited capacity to grow at the pace of enterprise business.
Don’t underestimate a CIO’s strive for more agility
Think it will be difficult to convince CIOs and IT professionals to give up their control over local systems and applications to a service provider? Think again.
True, a carrier is asking the CIO or network administrator to bet his job on handing over control of what was once the domain of the local data center. However, we are finding that enterprises are increasingly willing, even eager, to do so. Maintaining network and systems performance presents several challenges, including:
• Anticipating the need for additional hardware for specific applications.
• Purchasing, installing and maintaining that new hardware is costly and consumes IT limited resources.
Those challenges are made worse as more adopt a hybrid enterprise model with:
• A historically poor network visibility that is becoming a critical issue due to the “shadow IT” phenomenon and an increased use of Internet as a business network.
• A context of a much faster pace of rollout and adoption of new applications.
Debunking the myth of virtualization
With NFV, carriers can now share the compute required to run network functions for a given customer but also across multiple customers. NFV enables the delivery of services from within the service provider’s core network and points of presence to the customer’s premises using high-speed networks without the need to install hardware locally.
This leads to the question: Does NFV sharing resources and applications through virtualization necessarily lead to a waste of compute resources and to lag?
I firmly believe the answer is no – and I realize I must be 100% certain. Network and application slowdowns or outages mean a business is losing money as employee productivity falls and customers are unable to access your website and/or customer service resources, etc.
Virtualization has evolved to enable you to still operate in your own sandbox, allowing individual network functions and customers to obtain the resources they need. Virtualization and NFV present a valid alternative to hardware in most situations, especially when compute is shared across customers and as services are delivered from within the network core. In addition, NFV delivers elasticity, instantaneity and flexibility benefits that hardware will never be able to match.
Editor’s Note: The RCR Wireless News Reality Check section is where C-level executives and advisory firms from across the mobile industry share unique insights and experiences.
