|Issue:||Asia-Pacific II 2007|
|Topic:||A network revolution in Asia-Pacific|
|Title:||VP of Sales (APAC)|
Steve Robinson is the Vice President of Sales, Asia-Pacific, for Meriton Networks. Mr Robinson has 27 years of industry experience in sales, service and product development in Canada, the US and Asia. Prior to Meriton, Mr Robinson was the Senior Vice President of Avidia Networks at PairGain Technologies, where he was responsible for the development and global product management for DSLAM products. Prior to PairGain, Mr Robinson spent many years with Newbridge Networks (Alcatel), where he was Vice President and General Manager for the Canadian sales region and, later, Vice President and General Manager of the North Pacific sales region. Mr Robinson began his career at Mitel Networks in Canada. Steve Robinson graduated from St Lawrence College in Canada as an Engineering Technologist.
The Asia-Pacific region is a study in telecom contrasts. In the highly developed nations, Fibre-to-the-Home is becoming the standard for broadband service. In the less-developed regions, especially in hard-to-reach rural regions of the poorer countries, GSM mobile service is the local service benchmark. Over the next few years, however, the growth of bandwidth demand – first in urban areas to reach the relatively more affluent populations and then for mobile backhaul – will accelerate the growth of Ethernet-based NGNs.
The Asia-Pacific region consistently embarrasses – and leads – the rest of the world when it comes to the development and adoption of telecommunications services. This, undoubtedly, comes as a result of the region’s strong heritage in semi-conductors and consumer electronics, but also owes much to the tremendous appetite of Asian populations for bandwidth-hungry entertainment services and for ever-greater mobility. Taking the example of Fibre to the Home, FTTH, networks, the high-bar for next next-generation network deployment, Asia-Pacific has well and truly set the pace, with the other major global territories of North America and Europe lagging well behind. In this case, and with many others, a closer examination of the region uncovers a digital divide of its own. Over 70 per cent of APAC’s FTTH connections are in Japan, and the majority of the remaining connections are in Korea, Singapore and Hong Kong. By contrast, other countries, such as Malaysia and the Philippines, present a much different picture chiefly because of their radically different social and economic circumstances. New architectural approaches to the telecommunications network as a whole are changing this unique marketplace. Globally, many carriers are building the sort of next-generation networks, NGNs, typified by BT’s 21CN project in the UK. Fundamentally, these ambitious, long-term projects seek to simplify the carrier network to its very core to provide highly efficient, ultra high speed and capacity traffic delivery and greatly reduced operational, management and engineering costs. Carrier Ethernet transport – and next-generation networks Now, conventional networking thinking suggests that the way to achieve a next-generation network is to push the IP protocol into even the deepest layers of the network. The trouble is, IP isn’t just one protocol, it is an extended family of protocols, each one devised to meet some special set of circumstances that the original Internet Protocol was never designed to fulfil. Despite IP’s unprecedented level of customization to meet specific needs, none of the IP’s protocols has ever come even close to replicating the reliability and predictability needed at the heart of carrier networks. For example, there is a difference of several orders of magnitude between the restoration time of a carrier-scale synchronised optical network (SDH/SONET) and that of a large router network. So, despite shrieks of horror and accusations of blasphemy from the Internet purists, the world’s major operators are actively pursuing a more pragmatic line – the introduction of Gigabit Ethernet, GigE, which builds upon the proven reliability and efficiency of Ethernet. Today’s DSLAMs, Digital Subscriber Line Access Multiplexer, have GigE at the back end, as do some of the new mobile WiMax base stations. It therefore makes no sense to adapt Ethernet’s perfectly adequate basic framing and addressing format into something new and different just to convert it back at the other end. The new notion is that Ethernet protocols can be modified and enhanced to behave in a predictable, circuit-switched kind of way without sacrificing anything in terms of transport performance or efficiency. Carriers worldwide are migrating to next-generation networks, NGNs, to reduce their costs and support bandwidth-intensive services with quality of service, QoS, guaranteed by service level agreements, SLAs. They need a flexible, scalable optical transport infrastructure – especially in the metro network – that efficiently supports Ethernet and minimises operational complexities. Carrier Ethernet Transport, CET, is a new architectural design approach that recognises the inherent cost and manageability advantages of a circuit connection-orientated approach to carrier-scale transport. CET integrates enhancements to Ethernet Gateways/Tunnels, such as Provider Backbone Transport and Transport MPLS, multiprotocol label switching, with intelligent WDM, wavelength-division multiplexing, to allow carriers to use Ethernet to create direct, end-to-end, Ethernet ‘tunnel’ connections. Carriers are making these investments anticipating that their current infrastructures will soon be unable to support future growing services demands. High-income economies have a great hunger for broadband services that pushes average bandwidth per user requirements to more than 20 Mbps for both upstream and downstream traffic. As a result, the regional/metropolitan ‘core/edge’ portions of the network need abundant flexibility and capacity. CET solutions that meet these needs have emerged to support Ethernet switching within optical networks. Put simply, Carrier Ethernet Transport combines the simplicity and cost-effectiveness of native Ethernet with the reliability and power of WDM to deliver the flexibility, efficiency and cost savings that interconnecting with Gigabit Ethernet demands. As the network grows to include more and more endpoints, the need to switch traffic efficiently, end-to-end, across the entire network grows apace. CET embeds a number of key any-port-to-any-port network switching capabilities: Wavelength Switching, Sub-wavelength Switching (typically Gig E), and Carrier Ethernet Tunnel Switching. Furthermore, CET enables the separation of the service delivery architecture from the underlying transport architecture. Because of this, the transport network is not affected by adding, moving and changing – generally, the creation and modification – of services. Critically, CET optimises the hand-off of Ethernet traffic so there is less unnecessary consumption of ports on the switches. Indeed, this significant reduction to what some network planners call ‘head-end port explosion’ results in some of the biggest operational and capital expense savings – OPEX and CAPEX – achievable by carriers deploying a CET network. Carrier Ethernet transport architecture for NGNs Carrier Ethernet Transport comprises an optical layer of switching elements, which ensures end-to-end connection-oriented Ethernet paths that can be provisioned remotely and rapidly, and can be guaranteed with SLAs. The clear separation between the service layer and transport layer provides the best scalability and ease of deployment. By keeping the switching in the optical domain with its low latency, low jitter approach to switching Carrier Ethernet traffic, carriers can achieve defined QoS levels. CET enables end-to-end path management of individual GigE optical paths, with point-and-click provisioning and protection, while optimising fibre usage. Dealing with APAC’s digital divide Social and economic circumstances differ starkly between the most advanced and the least developed nations within the APAC region. Japan and Korea boast remarkable, truly world-leading, 100 Mbps+ broadband subscriber penetration rates by virtue of their high GDP per capita, concentrated urban populations and historic knowledge of – and enduring thirst for – hi-tech manufacturing and development. Across the region, carriers like PCCW, KT, NTT and – much more recently – Telstra have not so much embraced 21CN/NGN- type network overhauls as defined them. The high GDP nations have also benefited from economic stability and the sustained investment in a high-capacity fixed network infrastructure that this affords. Looking elsewhere in the region, at countries such as Vietnam and the Philippines, and eastwards toward India and Pakistan, low personal income and overall GDP have been responsible for restricting telecoms development. However, current economic growth in these areas is rapid, and they have comparatively young populations likely to demand entertainment services and communications applications. In much of the APAC region there are enormous rural populations with low personal incomes, the need to roll-out networks quickly and inexpensively to serve them has favoured the growth of mobile networks. The users in these regions typically use pre-paid service and their usage is controlled and sporadic, rather than US$20, or more, monthly contracts; ARPUs are among the lowest in the world. The reliance on mobile communications in these countries reduces the need for NGN architectures sporting the latest technologies, such as CET. Bandwidth loads on the network are well below international averages, and capacity is only troubled by a few thousand Internet cafés serving the surfing few. Accordingly, despite the possibility of some OPEX savings, telecoms providers, typically, have yet to reach the point where investing in NGN architectures, even for backhaul, has become a burning need. It is just a matter of time, however, before growing usage will force operators in these regions to consider deploying NGN solutions, including CET. NGN deployment in Asia-Pacific is following two different patterns; deployments in highly developed regions are quite different from those in rural regions where the circumstances are quite different. However, observers would be wise not to write-off the potential of NGN in developing regions. Low-cost GSM/3G is where the telecoms growth is happening in these places, and this kind of traffic immediately lends itself to SDH Ethernet networking. In Malaysia – perhaps the best example of a country crossing over from ‘developing’ to ‘developed’ in telecoms terms – a carrier there has recently issued an Ethernet over WDM tender. New WDM tenders are already including Gigabit Ethernet transport requirements. Furthermore, some developing countries have significant numbers of affluent subscribers who are capable of driving demand for NGN. India, for example, has around 1.5 billion people. If we assume that only ten per cent of the population is relatively affluent, there is a market there equalling Japan’s and dwarfing that of South Korea. Whether a carrier wishes to boost ARPU, Average Revenue Per User, or reduce operational costs, the challenge will be to cope with the growth of bandwidth demanded by new services. NGNs offer the possibility of turning growing bandwidth consumption into an opportunity for profit rather than a sticky, complex problem. The advantages that a CET architecture brings – scalability, flexibility and reliability and lower costs – can help make that scenario a reality.