Previous | Table of Contents | Next |
While ATM is expensive to deploy, it compares favorably with dedicated leased lines in a mesh configuration for intensive data communications. ATM offers significant performance advantages when compared to traditional data networking alternatives. However, ATM often requires total equipment upgrade, although Frame Relay and SMDS equipment vendors recently have smoothed that transition path.
Sample hardware costs include the following:
For the most part, carriers have limited their offerings to T1 and T3 bandwidth, with either CBR or VBR. The focus is on data applications, although some support for voice is beginning to appear at the time of this writing. Usage charges are applied by some carriers, while others such as Pacific Bell price the service on a flat-rate basis. Network costs generally include the following cost elements, which vary in their application and level according to carrier and tariff.
Example network costs include Sprints ATM services between New York and Los Angeles priced at $4,240 for T1 and $31,600 for T3. LDDS Worldcom charges $94,500 for T3 service with VBR between New York and Los Angeles; the T3 flavor with CBR is priced at $121,500 [11-46], [11-47], [11-48], [11-49], and [11-50].
ATM is intended to support any application which the contemporary mind can conceive. The driving force behind ATM development and deployment currently is that of data applications. ATM does a beautiful job of supporting point-to-point data communications, as well as point-to-multipoint data conferencing. In a high-speed, multimedia collaborative computing environment, ATM really shines as a workgroup solution.
As a backbone technology, ATM is a true solution to bandwidth bottlenecks. For instance, a number of carriers are deploying or have plans to deploy ATM to address the looming Internet meltdown. For instance, Pacific Bell announced in December 1995 its plans to deploy ATM to increase its Internet capacity from 15,000 to 600,000 e-mail messages per second using Stratacom switches. According to Pacific Bell, its users generate 40% of all Internet traffic [11-51].
Ultimately, of course, ATM is intended to support voice and video, as well as data and image over LAN, MAN, and WAN. At this point, however, the standards are lagging for voice and video. The ATM Forum and others are working overtime to address these issues, and there is no question that they will be resolved in time. ATM, after all, is the network technology of the future.
ATM LANs
ATM-based LANs deserve special mention. Although some users have deployed high-speed ATM in LAN switches, the driving force appears to be that of the 25.6 Mbps ATM standard. Promoted by the ATM 25 Alliance and later accepted by the ATM Forum, that standard promoted the deployment of ATM in a workgroup environment and in support of collaborative activities which are multimedia (voice, data, video, and image) in nature. Connectivity is provided via an intelligent ATM switching hub, which serves to mitigate congestion issues that would result from such traffic over conventional LANs. ATM LANs can run over existing cabling structures, including UTP and STP, although the switch requires equipment upgrade and new NICs must be installed in existing workstations. In direct competition with FDDI, 100BaseT and 100VG-AnyLAN, 25-Mbps ATM certainly does offer more direct compatibility with ATM in the WAN. As always, the huge installed base of Ethernet makes competing technologies a bit of a hard sell [11-37], [11-52], and [11-47].
Broadband ISDN (B-ISDN) was formally addressed in 1988 through the first set of B-ISDN standards from the ITU-T (I.121). These standards were formally revised in 1990, and continue to experience revision and augmentation. The ATM Forum, formed in 1992, builds on these standards through the development and promotion of specifications for equipment interfaces in the ATM network. As noted previously, B-ISDN builds on the services foundation of Narrowband ISDN (N-ISDN). However, B-ISDN is based on cell-switching technology, whereas N-ISDN is a circuit-switched standard. B-ISDN makes use of ATM as the backbone network switching and transport technology, with SDH/SONET as the backbone transmission medium.
As we discussed in Chapter 8, N-ISDN generally has been less than a stunning success. Deployment of N-ISDN varies from country to country, and within each nation on a state, province, and metropolitan area basis. Generally speaking, it is not widely available at the time of this writing. Even where available, it generally has not enjoyed great market penetration, due to high cost and lack of meaningful applications.
Opinions differ widely relative to the likely success of B-ISDN. Some analysts suggest that B-ISDN is dependent on the success of N-ISDN, while others suggest that it will leapfrog N-ISDN, becoming pervasive in the future world of the converged network. As B-ISDN prototype switches have been built and tested by all major switch manufacturers, the technology certainly exists. The uncertainty revolves around the value of the likely services to be offered, as well as the costs of deploying such a network.
B-ISDN is defined by the ITU-T as a service requiring transmission channels capable of supporting rates greater than the primary rate. The primary rate is defined in Primary Rate Interface/Primary Rate Access (PRI/PRA) as DS-1. In other words, B-ISDN is a set of services requiring broadband facilities at 45 Mbps (T3) or 34 Mbps (E3).
There are three underlying sets of technologies and standards that are absolutely critical to B-ISDN. First, SS7 is viewed as the signaling and control which will support B-ISDN, as it supports N-ISDN. Second, Asynchronous Transfer Mode (ATM) is the backbone network switching and transport technology. Third, SDH/SONET is the physical backbone network transmission technology.
Previous | Table of Contents | Next |