 | Issue: | Asia-Pacific II 2002 |
| Article no.: | 6 | |
| Topic: | Global Seamless Networks for Digital Revolution | |
| Author: | Takuro Ito | |
| Title: | General Manager, Optical Network Product Marketing Division | |
| Organisation: | NEC, Japan | |
| PDF size: | 24KB |
About author
Not available
Article abstract
The convergence of voice and data has created new opportunities in Asia-Pacific as the digital revolution, in the form of data and Internet usage, spreads from the USA. Global carriers are emerging to utilise new and high-capacity trans-oceanic networks that are so vital in the Asia-Pacific geographical context of island nations. Takuro Ito of NEC, proposes a dual-aspect network architecture solution comprising mesh networks (in a change from ring) and terrestrial and submarine integration of global networks.
Full Article
The explosion of Internet traffic in Asia-Pacific, following US trends, has triggered an increase in data traffic on a global scale. In turn it is driving the deployment of huge-capacity trans-oceanic networks and the emergence of global carriers. Increase in Data Traffic While voice communication growth is roughly proportionate to the world’s population growth, there is a growing trend in many countries for data traffic to exceed voice traffic. In the voice communication age, international traffic was limited. Language barriers and time differences prevented easy, real time, international communications. That was then. Now, written text communication such as e-mail lowers the language barrier and buffers time differences. Graphics, recorded sounds and videos are exchanged more easily than real-time voice. Deployment of Huge-Capacity Trans-oceanic Networks During the past few years many submarine networks have been laid in the Atlantic, to be swiftly followed by others in the Pacific and other oceans. The global deployment of these huge-capacity trans-oceanic networks depends on two key factors: dense wavelength division multiplex (DWDM) technology and competition between operators and carriers. Each submarine network can handle close to 80 Gbps (gigabits per second), depending upon the wavelengths used to transmit the data; by adding wavelengths, the trans-Pacific submarine networks can be upgraded to a total capacity of more than 2 Tbps (terabits per second). Emergence of Global Carriers Multi-national companies (MNC) with numerous offices worldwide need networks with high-capacity data channels and global coverage to communicate, internally, large amounts of information. To meet the demands of the MNCs, e-business, and residential markets, service providers (‘Global Carriers’), have been vigorously building up global networks connecting city to city. Today’s Problems Traditionally, global networks consisted of two separate types of networks, terrestrial networks within each country and submarine networks crossing the oceans, each being managed by a Network Management System (NMS). The Global Carriers now face several problems. Here, we focus on three: provisioning delays in establishing end-to-end communications channels; the trade-offs of network architectures with regard to reliability and efficient bandwidth use, and floor space reduction. Establishing End-to-End Paths As stated above, the speed of today’s business operations means that MNCs need data communications channels to rapidly connect branch, sales, and temporary offices. Network services need to be fast. The task of establishing working channels-it can take months today-needs to be accomplished in seconds. Today’s international networks consist of many parts. For example, if you need a communication channel between Beijing and New York, you might have to use the following, separately operated, networks. 1. local access network in Beijing; 2. backbone network in China; 3. submarine network between China and North America; 4. backbone network in North America; 5. local access network in New York. Within this sort of architecture, interfacing between the various segments is often quite challenging. It is easy to see that a considerable amount of time might be needed, often months, to establish the links for a global communication channel. Maintenance to keep the channels working can be complicated. Bandwidth Utilisation The SONET/SDH ring is the dominant network architecture in the terrestrial network. SDH (synchronous digital hierarchy) is an international standard for optical networking; Sonet (synchronous optical network) is the North American equivalent. These ring-based networks provide reliable services and the fast protection capability necessary for voice networks, essentially by duplicating the network. The extra network is for backup; it guarantees reliability. The efficiency of bandwidth usage in such a duplicated network can never exceed 50 per cent. Other architectures promise to be more efficient. Floor Space Reduction As a result of the exponential global growth of international traffic, global carriers are building large capacity networks and upgrading the capacity of existing networks. As capacity increases, though, the amount of telecom equipment installed at Points of Presence (POP), Landing Stations (LS), etc., goes up, as does the floor space requirement. This issue is of particular concern for the LS, which, as the interface between the land and sea networks, has both terrestrial and submarine equipment. The submarine equipment for such tasks as DWDM can be quite large. Global Seamless Network Solutions The Global Seamless Network concept, a promising network architecture, integrates terrestrial networks and the submarine networks using DWDM transmission technology. It results from the fusion of, essentially, two architectures: a global mesh network, and the integrated terrestrial and submarine system. Global Mesh Network The backbone is the principal component of a telecommunications network. It is the centre for all traffic that allows one to communicate with all others on the network. A backbone network has two essential functions. The first function is ‘transport’-it provides link by link connectivity. The second function is ‘cross connect’-it provides switching between the links of the transport function. Typically, in today’s networks, the transport function is provided by SONET/SDH ring networks and the cross-connect function is provided by digital cross-connect (DXC) equipment. Most such backbone networks are built by linking a series of rings together in a multi-ring topology. A mesh network is a form of a point-to-point network. In a point-to-point network each point or node on the network is connected directly to a neighbouring node. A mesh network, sometimes called a ‘fully-connected mesh network’, improves upon point-to-point topology by providing each node with a dedicated connection to every other node. This sort of network tends to be extremely reliable. When a direct link between nodes is down, or busy, traffic can be easily re-routed (imagine a fishnet) through other nodes to arrive at the destination. This sort of multiple connectivity also makes it easier to establish rapidly-even in real-time-fixed channels between any two points in the network By using a mesh configuration for terrestrial and submarine networks, we can create a global seamless network that provides efficient allocation of network capacity, extreme scalability and flexible route selection. Mesh network architectures are easily configured for traffic loads between any two points. All you have to do is to upgrade equipment at the nodes and allocate sufficient bandwidth or physical circuits (fibres, cables) for the traffic on the route-the whole network does not need to be upgraded. In the event that a link or a cross-connect fails, traffic can be re-routed and service restored according to the priority of the service that has been contracted. Higher-priority, perhaps higher-paying, service can be re-routed immediately by the quickest available route. Secondary traffic can be re-routed and restored by a more roundabout routing. Clearly, the seamless global network will provide a different class of service. Benefits of Global Seamless Network Solution Rapid Provisioning The provisioning process can be divided into three steps: · planning the circuits-finding the routes and circuits to provide the end-to-end connection the user demands; · physical provisioning-installing the devices in the previously identified circuit nodes adding the needed components (such as terminal equipment line cards) and physically connecting cables; · network testing. When the terrestrial and submarine networks are managed separately, the circuit-planning process for each network has to be conducted independently to provide for intercontinental connections. This means that three independent plans (terrestrial, submarine and interconnection) must be devised for end-to-end provisioning. When you do not know the entire network configuration, you need to plan the circuit sequentially-design part A, then part B that connects to it, and so forth-to maintain consistency of the circuit. Three sequential design processes are needed. The integrated management system of a global seamless network deals, on a unified basis, with both terrestrial and submarine network configurations. This results in a seamless database which enables end-to-end integration of circuit design and saves time in the provisioning process. The integrated management system of a global seamless network contains detailed information about the equipment provisioning at each node and available capacity information for each link. This eliminates the need for much of the trial and error and iterative planning cycles needed to provision today’s complicated links by rapidly finding appropriate circuits for each portion. Bandwidth Efficiency In trans-oceanic networks, line terminals or cross connects can only be located at specific points such as landing stations. As a result, some links may have to be redundantly deployed. For example: Flexibility for Partial Network Upgrade The most important advantage of the mesh topology is its efficient allocation of network capacity. With ring architecture, if you need to increase capacity between any two end-points, you must increase the capacity of the entire ring. This means that you must upgrade all equipment belonging to the ring and spend time on re-testing the entire network. On the other hand, mesh networks are quite flexible and easy to expand. You need only upgrade and equipment and test at those parts of a route where extra bandwidth is needed. Floor Space Reduction There are special repeaters that can be used to integrate submarine and terrestrial networks. We call them Terrestrial and Submarine Integrated Systems. A comparative analysis of space requirement for each option shows that such integrated repeaters enable us to reduce space dramatically. Importantly, this sort of integrated repeater also reduces considerably the power consumption at the facilities where these systems are installed. Conclusion The global seamless network solution for the digital revolution brings many benefits-rapid provisioning, bandwidth efficiency, flexibility for network upgrades, and the reduction of floor space and power consumption of the system. It will allow global carriers to build high capacity networks and to provide a wide variety of services at a competitive cost. The solution to 3G infrastructure buildout is, itself, a digital revolution.
