If you haven’t noticed, the world of mobile computing is going through a massive change. According to CTIA, wireless providers reported handling 2.1 trillion text messages in 2010 for the United States alone. This is up significantly from the 81 billion reported in 2005 and the 1.5 trillion in 2009. This also doesn’t even begin to account for the Facebook, Flickr, YouTube and numerous other heavy-data applications that today’s mobile users are tapping into. Consumers are no longer buying and using mobile devices with the primary use case of actually making a telephone call; some of these devices aren’t even capable of making traditional voice calls.
This is a significantly different service model than mobile carriers were built upon; this is digital data, not analog voice. Carrier networks that were originally designed to handle voice traffic have had to quickly cobble together additional network paths to handle data. Often times, these data networks have been implemented in much the same way that traditional voice networks were designed; treating all data as equal with minor segmentation between things like video and “all the rest.” However, even web-based data traffic is not equal. Some applications being used by users are heavy, frequent consumers, some are real-time and others are occasional, but carriers lack the ability to differentiate at the application level and thus, have no way to treat these different applications as unique services.
The business impact
The effect of all these changes presents itself in a reality of escalating costs to build and maintain the networks necessary to keep up with user demand (and simultaneously not negatively impact existing user experience), but decreasing average revenue per user (ARPU). Even the users who are willing to pay for increased bandwidth and data usage can negatively impact other users unless the carriers build out their network; causing increased capex and opex that is not made up with the revenue from those few heavy users. The highly competitive nature of the business also conspires to regulate how much a carrier can charge for data usage before users start looking elsewhere. Attractive alternative providers and overall user experience are driving significant subscriber churn which brings a whole new set of costs to the carrier’s business.
These are the tactical issues faced by the mobile provider and represent the reality of the business. As they start planning for the future, with LTE networks and the closely associated IMS, all-IP core, they have several strategic goals in mind. The first is a simplification and consolidation of the network to reduce capex and opex. The second is to find new ways to generate higher ARPU. Lastly, they are looking for ways to monetize their largest asset, the millions of subscribers on their networks.
The goal is to bring back the days of massive profitability by reducing costs, increasing revenue and creating brand new revenue streams.
Looking for a solution
While many mobile carriers may be loath to admit it, many of the challenges they are facing are quite similar to those faced by the enterprise over the past 10 years. The consolidation of networks to IP-based protocols and the tight integration of Internet based services and applications, or the “webification” of traditional enterprise applications, caused similar problems. When one considers that content providers like Facebook count some 500 million users accessing their application 24-hours a day, the scale and reliability requirements for some of these enterprise’s are also akin to the notion of “carrier class” requirements. Thus, looking at the successful solutions implemented in the enterprise is a legitimate place to start.
However, the carrier networks do have many unique features that make them much different from the enterprise; maybe not on a scale or base networking standpoint, but on a functional standpoint. Carrier networks require a much greater capability to monitor, measure, and document each user, the services they use, the amount of data they consume and the amount of time they are connected. In addition, they must also have a significant amount of information about what the user is provisioned to use, how much they have already consumed, where they are, the type of device they are using and whether they are using the home network or are roaming. This requires a significantly greater need for tight systems integration and dynamic intelligence that allows the network to not only deliver services, but to change how those services are delivered in real-time.
Service providers need a common IP architecture of reusable services they can apply, regardless of traffic type—voice, data, or video. The right set of services—large scale NAT, DNS, AAA brokering and scale, optimization, and security—becomes part of the converged IP architecture. Ideally, those services would maintain session context, provide visibility, and enable providers to easily act on the traffic regardless of the type. Such a system would require deep session intelligence, service integration management and programmatic system interfaces that provide links into back-end systems for subscriber policy information and billing, and adaptability to solve future challenges and overcome incompatibility of divergent systems.
>Service delivery Networking
That notion of an integrated, strategic point in the network is the basis for service delivery networking—taking tried and true technologies from the enterprise application delivery space but adding specific integration required for the unique demands of the mobile carrier.
Instead of duplicating common services, like security, optimization and scale for each type of traffic, service delivery networking focuses on creating a single point where these critical functions can be applied to all traffic. This strategic point of control also provides the most opportune point to integrate provisioning and billing systems as well. This consolidation and simplification can result in significant capex and opex savings, reducing the number of systems needed to manage subscriber traffic. There is also benefit in simplifying the data path for promoting increased performance, reliability, consistency and ease of troubleshooting.
Having a consistent method of service delivery also enables carriers to easily implement new services with minimal cost and effort. Being able to leverage existing capabilities as well as the wealth of contextual information gives carriers the ability to intelligently differentiate each service and create new opportunities for revenue. Creating rate plans that provide unlimited usage of individual applications (for example, unlimited Facebook access for $5/month) becomes a simple matter that allows subscribers to pay for the access that is most important to them and the carrier to increase ARPU by monetizing the applications that were once simply using the carrier as a transport.
Finally, having this centralized, intelligent and dynamic point of control also allows the carrier to finally monetize their subscriber-base. By integrating contextual information about the location of the user, the type of device they are using and their service preferences, the carrier can create revenue streams from outside advertisers or by up-selling subscribers. The ability to target highly qualified subscribers based on content preferences, device and/or location could be highly valuable to external parties. If a new video service were being launched, the ability to target heavy users of YouTube or Netflix with pop-up advertising could be invaluable. Or, the ability to pop-up advertising (and directions) to retailers who sell an item being searched for on the web could benefit both the retailer and the subscriber.
Service Delivery Networking is an architectural approach that lays the foundation for a unified solution reducing complexity, capex, and opex while simultaneously enabling the qui
ck and efficient implementation of new and differentiated services. This solution relies on deep session intelligence to provide true insight into the traffic and services being delivered; integrated services management to blend seamlessly with complimentary services and back-office systems; programmability to enable fast, efficient, and reliable policy enforcement; and adaptability to fit any service offering. Ultimately, the ideal solution reduces the cost of maintaining services, increased the reliability and can be leveraged to easily create new services that generate new revenue streams and higher ARPU.
KJ (Ken) Salchow, Jr. has been employed by F5 Networks Inc. for more than a decade where he has served in several capacities: SE, Security Architect, Technical Marketing Manager, Speaker and Corporate Spokesperson. He currently manages the technical marketing team who produces white papers, blogs, podcasts, videos and presentations that help define the role of Application Delivery within the enterprise architecture. Prior to F5, Ken worked for the Internet Services Group at Best Buy Corporation, the team that designed and architected one of the most successful and profitable clicks-n-mortar operations ever launched. He has owned and operated multiple businesses ranging from consulting to pen-testing to computer forensic and has at one time held CNA, MCSE, CCNP, Network+, CISSP, C|EH and CCE certifications as well as had formal training in Solaris, Check Point and wireless technologies. He has a AS in electronics technologies, a BS in information technology and is currently completing his MBA with a focus on IT. He currently lives in Minnesota and can be reached at k.salchow@f5.com.
Reader Forum: Service delivery networking: What is it and why is it important?
ABOUT AUTHOR