Previous | Table of Contents | Next |
Asynchronous Transfer Mode (ATM) was developed as a concept by the ITU-T as an outgrowth of ISDN standards which developed in the early 1980s. While the concept of Narrowband ISDN (N-ISDN) was intriguing, it soon became clear that the demand would develop for a wide range of bandwidth-intensive services which would be beyond the scope of narrowband (Nx64 Kbps) or even wideband (1.5 Mbps-45 Mbps) transmission facilities and circuit-switched connections. Given that assumption, which has since proved to be correct, the ITU-T cast an eye toward the future of broadband (> 45 Mbps) networking. Broadband ISDN (B-ISDN) was the conceptual result. Although B-ISDN still is very much in the future (N-ISDN has yet to reach its potential), its ATM network technology foundation is beginning to make a mark in the market.
The first ATM network in the U.S. was a testbed gigabit network known as the National Research and Education Network (NREN). Sponsored by the U.S. government Advanced Research Project Agency (ARPA) and the National Science Foundation (NSF), the project began in 1990. There currently are a number of such testbed networks in the U.S. In Europe, the Research for Advanced Communications in Europe (RACE) project 1022 was initialized in 1987 to demonstrate the feasibility of ATM. RACE is sponsored by a consortium of carriers, end users and universities. The R1022 ATM Technology Testbed (RATT) has exceeded expectations. RACE project 2061, also known as EXPLOIT, is a more recent RACE project intended to prove the viability of Integrated Broadband Communications (IBC) in the European Community (EC). In Japan, Nippon Telephone and Telegraph (NTT) has embarked on a project with the cooperation of manufacturers to replace the existing network trunk structure with ATM by 2005 [11-36].
Asynchronous Transfer Mode (ATM) is a fast-packet, connection-oriented, cell-switching technology for broadband signals. ATM is designed, from concept up, to accommodate any form of informationvoice, facsimile, data, video and imageat broadband speeds and on an unbiased basis. Further, all such data can be supported with a very small set of network protocols, regardless of whether the network is local, metropolitan or wide area in nature. In terms of user access rates, ATM currently operates at access speeds of DS1 and DS3, with much higher speeds planned in the future. The backbone transmission rates are DS3 and OC-1 at a minimum, and often at OC-3. A 25.6 Mbps ATM LAN standard was more recently and reluctantly approved by the ATM Forum, with strong encouragement from the ATM 25 Alliance. [11-37]. ATM is highly scalable in terms of both bandwidth and geographic reach.
ATM can be distinguished from Frame Relay in that ATM is a backbone network technology, whereas Frame Relay is an access technology. ATM also has application on-premise through LAN switches, hubs, routers, and the like. PBXs eventually will be equipped with ATM switching matrices, as well. Through the 25 Mbps standard recently adopted, ATM also is finding its way to the workstation.
ATM data can be of three types. Constant Bit Rate (CBR) traffic, such as T-carrier, requires the presentation of time slots on a regular and unswerving basis. Variable Bit Rate (VBR) traffic, such as compressed video and bursty LAN traffic, requires access to time slots at a rate that can vary dramatically from time-to-time. Available Bit Rate (ABR) traffic, such as bursty data, can deal with time slot access on an as-available basis. ABR traffic also is known as best-effort ATM. As we will discuss later in this chapter, all earlier classes of service are now specified as CBR, VBR and ABR in the ATM world; in each case, specific Quality of Service (QoS) parameters are defined.
In any case, and as the data are presented to and accepted by the network on a start/stop basis, the transmission mode is asynchronoushence the termAsynchronous Transfer Mode. The transmission facility, of course, is highly synchronized; the ATM switch and all other network elements are synchronized with the pipe, as well.
The user data are sent to the network over a digital facility. At the workstation, router, or ATM switch, data are organized into 48 octet cells. Each cell is prepended with a header of 5 octets and multiplexed, contending for access to a broadband facility, ideally SONET in nature. As the ATM cell is 53 octets in length, with 48 octets of payload and 5 octets of overhead, it is reminiscent of SMDS. SMDS, as we discussed earlier in this chapter, actually is a variation of ATM. While SMDS provides an immediate solution for certain applications, it likely will be relegated to niche applications in the future. ATM is intended to be the ultimate and complete network transport solution, supporting a theoretically infinite range of services via B-ISDN.
The advantage of the small cell size is that it can accommodate any form of datadigital voice, facsimile, data, video, and so on. The fixed length of the cell offers the network switches the advantage of predictability, as compared to a variable length frame. These two considerations yield decreased delay, as data move through the switching systems and across the transmission links in frequent little blasts. Long, and especially variable, frames occupy the attention of the network for relatively long periods of time, causing delay as other data would have the effect of waiting their turn.
ATM is the first network technology to offer true Bandwidth-on-Demand, as the bandwidth can be varied during the course of the call [11-3]. True enough, other services offer bandwidth which can vary with each call, but none can offer the ability to adjust the amount of bandwidth required to support a call once the call is established. For example, a high-quality videoconference might require 1.544 Mbps (T1) capacity as a rule. Yet, with the sophistication of contemporary compression techniques, that call might require much less bandwidth much of the time. Once that call is setup, a full T1 is dedicated to it, regardless of the actual bandwidth requirement moment-by-moment. ATM is not so rigid, as it can adapt dynamically to the bandwidth actually required.
As is the case with Frame Relay and SMDS, ATM networks do not provide for error detection or correction. Neither do they provide for protocol conversion. Rather, these responsibilities are shifted to the end user. The advantages are increased speed of switching, elimination of associated delay, and reduced cost because the ATM switches require less memory and processing power.
Previous | Table of Contents | Next |