TV Snow and Rain Fade

xChange Magazine
by Tara Seals
November 30, 2006

Weather and telecom services have a few things in common: both are subject to change and can degrade quickly. In weather, we can interpret the patterns of upper-level high and low pressure systems to guesstimate which coat to wear in the morning. But service providers need to go beyond predictions to assurance because if next-generation services don't live up to their quality forecasts, operators could find themselves permanently out in the cold.

As core networks are converging into a single multiservice transport layer, network operators will be able to deploy innovative, software-oriented, IP-based services on a common packet infrastructure nearly as fast as marketing executives can envision them. Operators are banking on these services to provide new revenue, monetize the capex they've put into network transformation and outmaneuver the competition.

"While the operators are still selling broadband access, they recognize that the access is no longer the primary service," says Jarrod Siket, vice president of marketing at Tollgrade Communications Inc., which provides remote test and management. "Yes, it still serves as the method for largely unprofitable high-speed Internet access, but it now has become the link to a series of highly profitable, network-based IP services. These include VoIP, IPTV, online gaming and others."

However, adoption of these services won't happen if the end user doesn't receive a quality experience. "There is a fiercely competitive battle to gain customers away from the entrenched provider [for things like television]," says John Burnham, vice president of marketing at Brix Networks, a performance management vendor. "And there are economic imperatives driving innovation. You have Google and others entering the game now. And there is a correlating inflection point in service assurance — you simply cannot risk churn or erosion due to quality problems."

But spotting quality storms is a complex task. To service-assure the next-gen network world, service forecasters have a much larger picture to consider. In the past, a service went hand-in-glove with the specific network that carried it, and service quality was a function of what happened in the network. Monitoring suites and troubleshooting platforms were placed in the network to ensure the quality of experience for the end user. Now, with a multiservice network in place, "the source of that problem may be in the home configuration, the broadband access network or service, the metro Ethernet aggregation network, the regional IP network, or the content and service delivery points — a softswitch, media gateways, IPTV servers, etc.," says Siket. "In order to quickly verify, isolate and repair these problems, the operators are unifying their operations procedures and tools that provide them visibility into all of the layers of the network that may affect the services they provide."

The result is a migration of quality assurance beyond the network, to encompass a carrier's IT infrastructure and front-office applications. "In the last three to five years, operators have moved beyond network fulfillment into service fulfillment, creating an OSS to deliver as many services as they can muster," says Sanjay Mewada, vice president of strategy at NetCracker Technology Corp., which offers a customer analysis module that helps providers understand the customer experience by correlating alarms from the network with service definitions and characteristics. "So services don't just live on the network anymore. It's not enough to monitor the network, and you can't simply extend fault management and test and measurement platforms. You still need information from the network on degradation and faults, but you have to marry that to which service is affected in which configuration for which geographies and which subscribers. That information lives in the OSS — the service fulfillment system. They need to understand the services, the SLAs, the profiles and the subscriber data, in order to understand the impact on the customer."

This move to a holistic, unified view becomes more critical as time-shifted and geography-shifted services become more common. "Some services also now can have short lifespans, such as on-demand content," says Mewada. "When someone pays for an on-demand soccer match via IPTV, they expect it to be high quality. So, the service assurance challenges become much more complex and demanding than in the traditional world because everything goes to real time."

In the event of degradation in the past, providers would get a phone call from a subscriber, then go out and fix the problem within hours or sometimes even days. "Now, two hours is too late — indeed, two seconds might be too late," says Mewada. "You have to be very aware and capable of managing the experience on a real-time basis. As soon as there's a problem, you need a system that proactively alarms everyone involved and takes protective measures, whether that's adding more bandwidth or reassigning circuits before the threshold gets to that inflection point."

The focus on the individual subscriber is another new atmospheric condition. "There have long existed methods that provide guesstimates to what a user might be experiencing," says Chris Girardeau, CEO at QoSmetrics Inc., which provides end-to-end network performance monitoring and diagnostic products for triple-play services. "These are no more than inferences made from findings in the core network. What is best to achieve successful end-toend assurance is a method that can pinpoint a user's problems. A service provider has to be able to look directly into the home for problems, not just make best guesses from the core."

Such an approach is critical if providers are to weather the service innovation storm and meet larger business objectives such as revenue growth, customer retention and growth assurances, he adds. "Proactive, end-to-end monitoring is necessary, and it must be continually repeated throughout the life of a service — from the moment one decides to assess network capabilities to deliver new IPTV services, for example, on through to supporting customer SLAs," explains Girardeau. "This requires tools that measure deep into the network and not just tell you whether a service is simply up and running. The right tools require deep drill-down capabilities to pinpoint discrepancies right down to the device or application in the core, or in the home."