Making broadband a boon

When preparing to deploy broadband services and all their promises, service providers should remember the real revenue driver: the end customer

Unleash the potential of the network!" That is the call that can be heard from one end of the broad-band industry to the other, from ISP to equipment manufacturer, upstart competitive local exchange carrier to behemoth incumbent LEC. Until recently, the call drowning out all others was "Deploy the network!" And deploy they have.

The dream of broadband, always-on Inter-net access for residential and business customers is now a reality. Statistics confirm this. Users subscribe to broadband services at an estimated rate of more than 300,000 new accounts each month and are expected to exceed 500,000 per month by the end of the year.

Of course, the pattern of end-user growth is not limited to the broadband arena. Internet traffic, including analog connections, is growing at 100% to 200% annually. This means that approximately seven out of 10 people or businesses in the U.S. alone - more than 400 million people - will use the Internet by 2003. It is clear that not only is the Internet an increasingly popular, diverse medium, but with every passing day, end users are becoming more sophisticated. They are, in essence, expecting more from their connections.

Now that deployment of broadband networks is reaching fruition, the resulting increase in demand for services has created new challenges for providers. Users are multiplying like wildfire, and demand for bandwidth by 2005 will be an estimated 200 times today's demand.

At first, the onus seems to be on providers to grow their customer base while maintaining a high degree of service quality. A noticeable decrease in service value would clash with consumers' heightened expectations of broadband service. Consumers already expect broadband to eliminate the slow performance and uncertainties associated with dial-up access.

Therefore, unless ISPs can find a better way to manage and guarantee the available bandwidth within the last mile of service, broadband will offer minimal advantages over dial-up as traffic grows.

We see this early attempt to manage service delivery in a flurry of cross-industry activity. ILECs, CLECs, building LECs (BLECs), interexchange carriers, ISPs, application service providers (ASPs) and IP service providers, among others, are selling services to each other in an effort to offer the broadest territorial and service coverage. While all of them are courting ISPs to accelerate customer subscription, many have started engaging directly or indirectly in selling services to end customers, primarily through the acquisition of an ISP. Meanwhile, ISPs are not idle. Most are reselling services from several facilities-based CLECs and now offer different broadband technologies such as DSL and wireless to their end customers. In fact, several ISPs are expanding their services to become ASPs, IP service providers and BLECs.

And the jury still is out on which ultimately will own the end customer.

As if that weren't enough, adding to these challenges is the impetus to differentiate; that is, to maximize the network's revenue potential through value-added services such as voice over IP, video-on-demand, multicasting, distance learning, videoconferencing, application hosting and IP faxing. However, to guarantee the quality of these services, ISPs must implement comprehensive service management capabilities throughout the network that will ensure data delivery and enhance latency-sensitive services (Figure 1).

Successful management here Because of the forces of technology development and service demands, DSL, cable and wireless networks have come into existence through what is, by and large, a piecemeal approach. There simply was no precedence for the evolution of IP technology. We now have reached the point where technology solutions can, and should, be implemented to match the dynamic nature of IP-based networks.

As the new value-added services mentioned above emerge, the fat pipe available today to the subscriber for data only will need, in fact, to become one that is shared among voice, video and data services in what the market calls convergence. Because the maximum bandwidth to the user often is limited by technical considerations such as the distance from the subscriber to the central office, once voice, and particularly video, have grabbed the bandwidth they require, the pipe remaining for data will become constricted to unacceptable levels. This will become debilitating unless the network is prop-erly managed and optimized.

It is this ability to manage that leads to a paradigm shift. The current paradigm of DSL service is Internet access offered to a residential or business subscriber at a flat rate. The new paradigm of value-added services is based on the dynamic delivery of IP services to the IP node (desktop or appliance) and a per-usage based billing scheme. And as such, this new paradigm transforms what currently is described as a pipe into individual services delivered to the desktop.

Therefore, provisioning, bandwidth use, user and service prioritization, quality of service (QOS) and usage-based billing of the available pipe to the user (vs. the demarcation point) must be carefully and efficiently structured to integrate seamlessly with the delivery of IP services to the individual user. Unfortunately, current architecture, which systematically deploys ATM, cannot accommodate end-to-end transmissions. And although it is adequate for basic broadband access, it will not be scalable over the long term as services and diversity of interconnections will create a labyrinth of virtual circuitry that is complicated and costly to provision.

Challenges in the industry should not translate into challenges for the end customer.

Provisioning, for example, is one of the most pressing problems that faces service providers. Service providers are having difficulty tracking customers and providing timely service to them. Increased demand will only exacerbate this situation. A key driver for future competitiveness for broadband carriers is to lower their high operational costs. Moreover, how will a service provider manage and adjust service levels of so many customers as demand skyrockets?

Such dilemmas and history speak to the wisdom of distributing IP intelligence throughout the network. To use DSL as an example, we can see an immediate need for IP intelligence at the customer premises. It should enable the dynamic delivery of IP services; at the DSL access multiplexer - when permanent virtual circuits and switched virtual circuits are terminated there; at the ISP point of presence for ISP driven IP; and at the central point of aggregation at the simplest level.

In a sentence, a service provider's survival depends on customer ownership, which, in turn, is a function of customer service and satisfaction. Therefore, service providers must become champions of service quality - differentiating levels of service and enforcing policies attuned to each service level agreement (SLA) - and management, focusing on bandwidth allocation and subscriber services.

This shift toward ubiquitous IP infrastructures already is being felt throughout the industry. However, the sooner equipment manufacturers can integrate IP-based solutions into their products, the faster service providers can differentiate themselves and boost profitability (Figure 2).

Unleashing the network When taking stock of the areas that service providers need to address in the development of technology solutions to maximize broad-band networks, a list of characteristics forms. Service providers must be able to monitor and evaluate traffic patterns. They must be able to identify and deal with abusive users. They must be able to manage bandwidth to server farms and application servers. They must issue and validate SLAs, as well as develop and manage tiered services.

From these characteristics, a formula for a solution begins to take shape. A solution must permit end customers to select service seamlessly and automate all steps of the provisioning process. It must guarantee the quality of the subscribed service and enable per-usage billing for the services that require it. And it must do all this in a cost-effective and scalable way.

The first killer app will likely be data/voice/video convergence, giving way to a new era of broadband service management needs. Thus, a comprehensive suite of advanced bandwidth management and IP QOS technologies, combined with subscriber service management systems that cover the front end all the way to the back office, are the answer to challenges arising in this new era.

At the heart of bandwidth and service management is the desire to maximize the end users' satisfaction and value. The value-added services and guaranteed levels of service establish the foundation for meeting customer expectations. Additional services afforded by usage analysis will allow service providers to exceed expectations as they venture into new frontiers of broadband capabilities.

The multi-tiered subscription rates that accompany the service policies will enable service providers to charge subscribers for flat-rate tiered services first, then for volume (data sent), bandwidth (speed) and services (usage). Hence, higher revenues can be generated compared with the current commodity-based pricing model. And so, for those in the broadband industry aggressive enough to deploy IP-based solutions throughout their broadband network, alongside their ATM network if necessary, the boon will not only be in customer retention but will obviously result in the increased profitability that Wall Street now demands.