Home Latin America II 1996 Multimedia for electronic commerce

Multimedia for electronic commerce

by david.nunes
Oscar CardaldaIssue:Latin America II 1996
Article no.:10
Topic:Multimedia for electronic commerce
Author:Oscar Cardalda
Title:Not available
Organisation:Stratacom
PDF size:36KB

About author

Not available

Article abstract

Oscar Cardalda describes ATM Technology and how it will benefit the way in which business will be carried out in Latin America in the future.

Full Article

The past decade has witnessed a revolution in the way information is stored, processed, and communicated. Centralized hierarchical computing and communications has given way to distributed processing, peer to peer communications, and client-server computing. ATM Technology Drivers As a result of these new networking capabilities, organizations have changed the way they do business. It is no longer necessary to assemble people within one location or even within one company to achieve business objectives. New tools and networks make it possible to move information wherever it is required, and collaborative applications enable users in geographically dispersed locations to work together productively. At the same time as companies are changing their internal computing structure, they are also deploying communications links beyond their own corporate locations to other companies and external networks. The increasing interdependencies of today’s economic world, coupled with the need for immediate access to information, are creating the need for on-line connections to the company’s suppliers, customers, and partners, as well as to communications gateways, information services, and the worldwide Internet. This trend towards conducting critical business functions via communications lines is often referred to as “electronic commerce”, and it is changing the face of the marketplace for many businesses. These new modes of work, involving distributed computing and inter-company links, have resulted in increasing demands on the capabilities of the wide area network, in terms of both high bandwidth and any-to-any connectivity. If these communications were to be provided using dedicated links, they would be prohibitively expensive. Shared bandwidth services, such as frame relay and ATM, provide the required performance and logical connectivity that users need at a much lower cost. The resultant approach to communications is virtual enterprise networking, a term which reflects both the evolution in the business model and the emphasis on logical connectivity. With virtual enterprise networking, users achieve access to the resources they need and the network provides the connection path and allocates the necessary bandwidth. Clearly, the time is ripe for a new network infrastructure optimized for virtual enterprise networking. The stage is set for ATM. The New Requirements Whether you’re a public, private, or hybrid network provider, the following requirements must be taken into account: 1. The network must be managed efficiently and cost-effectively. Regardless of the traffic mix, equipment must use bandwidth capacity efficiently. 2. The network must provide cost-effective access to the ATM WAN. Users aren’t going to throw out their existing applications. The WAN must integrate existing network services and requirements. It must support multiple services and interfaces at various speeds, as well as provide a migration path from low speed services to ATM services. And all services must interwork transparently, no matter what type of device is used to access the network. 3. The network must be able to deliver maximum performance and reliability. Traffic management, congestion management, network management, and multiple class-of-service capabilities are critical to the long-term success and performance of the network. Why ATM? ATM eliminates the inefficiencies of TOM technology by dynamically sharing network bandwidth among multiple logical connections. Instead of slicing the bandwidth into fixed-size channels dedicated to particular connections, ATM uses the entire bandwidth of a WAN trunk to transmit a steady stream of small fixed-size cells. Each cell has an address to identify it with a particular logical connection. Connections needing more bandwidth are simply allocated more cells, and when a connection is idle no cells are generated, and hence no bandwidth is consumed. This feature makes ATM the ideal switching technology for the bursty traffic types that will dominate the networks of the future. Computer-to-computer applications also demand very large amounts of bandwidth. ATM also addresses this requirement operating at speeds into the gigabits-per-second range. Fixed-size cells with a simple and consistent format enable switching functions to be performed in hardware, and also result in reduced serialization and queuing delays. ATM is multimedia. Virtually any type of information can be converted into cells for transmission over an ATM network. In fact, traditional traffic types like voices are quite bursty in nature. During an average conversation, each party speaks only about 40% of the time. By generating cells only when speech is present, ATM can reduce the requirements of a voice connection by half. ATM Applications: ATM Market Segments Not long ago, the ATM market was seen as a single market for ATM technology. However, based on much experience and research, a more segmented view of the ATM marketplace has emerged. There are three discrete segments in the ATM marketplace with three different sets of requirements. These segments are: the ATM LAN, ATM WAN, and ATM Central Offices (COs). Communications companies make very different design decisions depending on which segment they are targeting. The ATM LAN The ATM LAN segment is the low end of the market for ATM products. The key concerns of the suppliers of LAN technology are high throughput speeds and connectivity. In this segment, bandwidth is essentially free because once a building is wired for fiber, high bandwidth is essentially there with no incremental costs. Vendors designing for this segment use high speed interfaces and don’t worry about congestion control or bandwidth optimization. Local area ATM-based equipment includes ATM switches, routers, hubs, bridges, and workstations. When the individual workstations and applications attached to the LAN become faster than the LAN itself, then the number of users that a single LAN segment can support goes down. This creates the need for users to aggregate their data traffic onto intelligent ATM Hubs. Hubs provide high speed concentrated access for many users to a shared resource (i.e. a server or router). ATM routers or hubs are access devices that accept multiple routing protocols, and convert them into ATM cells for transport over the ATM WAN. A local ATM switch then interconnects the router or hub to the ATM WAN. ATM-capable workstations, servers, and PBXs can also be connected to the local ATM switch, forming a high-performance work group. From 1997 to the end of this decade, users will begin another premises investment cycle, which will be focused on extending ATM services to the desktop via the ATM UNI interface. Multimedia applications will fuel the demand for native-mode ATM interfaces. However, for many users, ATM to the desk-top is still too costly, although the price for an ATM interface card is dropping rapidly. There is currently only a small number of native ATM interfaces available for end user systems. The ATM WAN StrataCom focuses on the ATM WAN segment, which includes both public and private WANs. In fact our business today is split about 50/50% between public networks and private networks. Here, bandwidth is not free – our customers worry a lot about bandwidth costs. Therefore designing bandwidth-optimized products is very important. For example, using efficient coding and compression schemes reduces the amount of required bandwidth. Offering multiple services is also very important because the ability to translate all traffic into cells enables users to cost-justify a high bandwidth infrastructure. It is as a backbone switching technology that ATM excels. The asynchronous and multi-media capabilities of ATM means that it can be used to carry many types of traffic across the WAN. The ATM UNI is just one of several ATM interfaces/services that can be provisioned on an ATM switch. Others include frame relay, SMDS, and circuit emulation. Each of these services offer superior performance when provisioned on an ATM network. The ATM Central Office The high end of the market consists of ATM Central Offices. These are very large systems that must support a large number of user connections and a variety of services. Where ATM LANs might exist at 1 to 2 Gbps, and ATM WANs at 5 to 20 Gbps, ATM central offices are designed to be optimized at 20 to hundreds of Gbps. Although CO switch vendors may talk about smaller, “entry-level” versions of these units, they are still designed to be optimized for larger configurations. Typically they contain only ATM native interfaces. Like ATM LANs, bandwidth optimization isn’t a key issue for the CO segment because these units are installed on fiber rings. The issues that drive this segment are the ability to support multiple services, scalability (which ensures the ability to grow for ten to twenty years without outgrowing the platform), management, and billing issues. ATM Market Challenges WAN Connectivity One of the biggest challenges for ATM is in the WAN. ATM has been defined at speeds of 45 Mbps, 100 Mbps, 155 Mbps, and up to 622 Mbps in the ITU-T standards. While these speeds might be appropriate for the LAN connectivity, the majority of user locations can’t justify the high bandwidths defined for ATM connections – 56 Kbps WAN connections still outsell T1/E1 connections by ratios ranging from 3:1 to 10:1. Even five years from now, only about 10% of the locations will be able to justify T3/E3 speeds and costs. Thus the demand for scaleable ATM-based switching platforms that support multiple access interfaces at narrowband, as well as broadband speeds, are required for early adoption of ATM technology. To achieve high speed LAN/WAN connectivity, some users are building hybrid networks that are a combination of private data networks and public WAN services. For others, it is more cost-effective to access public carrier WANs to take advantage of services such as frame relay and ATM, rather than building private networks. The increasing demand for public frame relay services in turn provides the carriers with a solid economic incentive to invest in ATM WANs. The New Service Dilemma The problem of ATM supply and demand can be summarized as follows. We have carrier networks optimized for voice communications, because that has been the traffic type most often carried over geographical distances. We have local area network structures optimized for data, because it was data applications running on LANs that justified user investments in LAN technology. We have a new service concept, ATM, which, while inclusive of both voice and data traffic types, can’t be delivered over either the existing local network infrastructure or the existing carrier network infrastructure. With any new service, one begins with no network and no compatible CPE. Carriers try to seed the demand for the new service with the deployment of pilot network services for early adopters of the technology. However, since there is not a lot of traffic on the network, carriers don’t invest heavily in network expansion. This keeps costs high and ubiquity low. Users continue to “wait and see”, content with putting off commitments to new CPE and services. It becomes the “you move first” standoff. This dilemma is, in fact, what has kept ISDN stuck on the launching pad for ten years. If carriers are unwilling to make the necessary investment in ATM without pre-existing demand, and corporations are unwilling to invest in new broadband applications and access equipment without pre-existing ubiquitous and economical services, how will ATM emerge? The answer lies in creating ATM-enabling solutions. Why StrataCom? Ovum, one of Europe’s foremost research and consultancy companies that specializes in leading edge computing and telecommunications markets, makes this statement in their 1993 market report on ATM switch products: “The rapid development of the frame relay market and the subsequent eclipse of SMDS is based largely on a single product – the StrataCom IPX cell relay switch. It has become the ‘plug and play’ platform for offering wide area data services. It provides an irresistible combination of flexibility in handling multiple interfaces and more efficient use of wide-area transmission capacity.” Conclusion StrataCom has implemented all of the following solutions: · StrataCom’s scaleable multiband ATM platform allows you to take full advantage of the performance benefits of ATM. We support all of your existing services on one integrated platform, and provide full interworking among them. · All service interfaces and features comply with current standards for ATM switching · systems. · Our advanced “value-added” features ensure that your network is reliable, efficient, and “future-proof”. StrataCom offers sophisticated solutions for network congestion management and control, automatic routing and re-routing around failed connections, multiple classes-of-service, and fairness among users.

Related Articles

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More