Everything about how we communicate has changed over the past decade. For written communication, we’ve gone from letter writing to email and from email to social media postings. For voice communications, we transitioned from fixed line phones to mobile phones to texting and video calls over the Internet. For data communications, we’ve gone from computers to laptops to tablets and smart phones – let alone the ability to connect to the network anywhere, at any time and in any place. Furthermore, entertainment services have moved from analog to digital and single screen to multiscreen.

To further emphasize the impact of these changes – many of these services are moving from household based to individual based services – further challenging network operators to adapt quickly.

Nothing is more indicative of this change than the migration of access lines from fixed to VoIP and mobile. As shown for the U.S. market, fixed access lines have rapidly declined as demand shifts to mobile and VoIP – with VoIP now representing 26% of access lines. And, according to the latest statistics (2011) from the U.S. Department of Health and Human Services, more than one-third of all households are now wireless only.

Network operators have acknowledged that their future is dependent on their ability to build flexible, intelligent, IP-based networks that will enable them to deliver converged communication and multimedia services that customers demand. These new networks must be able to allow for faster introduction of new services; operate at reduced costs while increasing efficiencies, and providing customers with control, choice and flexibility in their services.

Although operators have been transitioning parts of their networks towards IP over the last decade – the approach has been fragmented – resulting in both high capital and operating costs with significant redundancies and inefficiencies.

The goal of the All-IP network is to completely transform (“to change in composition or structure”) the 100+ years of legacy network infrastructure into a simplified and standardized network with a single common infrastructure for all services.

Operator Strategy will vary

The starting point for each operator will likely be different as the dynamics of each market vary. Operators with extensive legacy infrastructure will need a different approach than those with more recent network investment (e.g., North America versus India). Additionally, the regulatory environment will play a key role in the migration strategy for many operators. However, the common thread for all operators will be the convergence of the voice and data networks into a single IP network.

Over the past decade many operators have successfully implemented network migration strategies based on NGN and IMS, with BT taking a precedent setting approach with its 21CN initiative back in 2004. Unfortunately, the dynamics of the telecommunications market, particularly in the broadband, mobility and application segments were drastically underestimated by the architects – but they did get a few things right – especially their view of Ethernet as a change agent.

Thinking outside of the box

To anticipate future change in network demands and services, Deutsche Telekom is taking a more radical approach with its recently revealed TeraStream architecture, a system that combines cloud and network technology. The simplified architecture will comprise of two types of routers: R1 and R2. The R1 routers are used for customer aggregation and all policy, and the R2 combine the functions for core, peering and datacenter switching/interconnect. The R1 and R2 routers are connected by an optical ring that forms the core of the network, while the core transmits Ethernet frames carrying IPv6 traffic allowing policy processing for different services through a single IP lookup.

Key elements of the TeraStream architecture include an all-IPv6 streamlined routing architecture; fully converged IP and optical layers with 100G coherent optics tightly integrated with the routers; integrated cloud service centers, enabling virtualized network services and applications for rapid service innovation; programmatic interfaces aligned with the software-defined networking (SDN) architecture for real-time automation and OSS; and customer self-service management capabilities.

Prior to this new architecture, the BSS and OSS functions were highly fragmented resulting is long innovations cycles for the introduction of new services. The TeraStream architecture retires the legacy systems and provides clear distinction between OSS and BSS functionality while allowing services differentiation towards customers; instant provisioning; instant change of access features; a reduction of products innovation cycle from 2 to 4 years to less than ½ years; no latencies; and significant cost advantages.

What differentiates this concept from other All-IP implementations is the fact that all the services, including the traditional Telco services (voice, IPTV, Internet access) will be delivered from the Cloud as opposed to the network.

DT is currently in a one-year pilot trial of its TeraStream architecture in Croatia through its Hrvatski Telekom subsidiary. In this market, DT plans to have 100 percent of its network migrated to All-IP by 2014/2015.

Regulatory Impact

A big question for many markets will be how much regulatory policy will impact operator strategies for All-IP. In the U.S. market, AT&T has stated that current regulation has had an “investment-chilling” effect on infrastructure investment and requests the FCC to take action now to facilitate the transition from TDM-to-IP and prevent stranded investment in obsolete facilities and services.

In Europe, concerns remain about competition and the impact these network transformation will have on unbundled local loops while balancing the goals of the Digital Agenda for Europe.

Regardless, the market is changing and fast. Yogi Berra may have said it best: “If you don’t know where you are going, you will end up somewhere else”. The move to All-IP is not an if, but more of a when and it is certainly closer to sooner than later – perhaps by the end of this decade.