Report - ATM networks (Asynchronous Transfer Mode)

By Assad Chaudhry 1995

The Report is also available for Download (PDF Format). Any part of the report should not be copied or used for commercial purposes without the consent of the writter. Students and non-commercial boddies are welcome to use this report for learning purposes.

The report will contain an explanation of the Asynchronous Transfer Mode (ATM). It starts with explanation of ATM basics and will provide an insight into the rapidly changing area of advanced communication technology.An understanding of the deployment of ATM in solving today’s problems and Realistic understanding and providing solutions for the future will also be contained in the report .

Finally , the report will discuss ATM standards and their implantation including few case studies.

 

Telecommunication, video conferencing and fast networks are becoming more popular than ever before, which is giving rise to bandwidth requirements. Today’s existing networks are simply not suited to the bandwidth demand placed on them. In result ATM (Asynchronous Transfer Mode ) was introduces to full fill such demands.

Fig.1.a. Bandwidth Requirement Ref: K-Net ATM Guide
ATM delivers important advantages over existing LAN and WAN technologies, including the promise of scaleable bandwidth, performance and guaranties Quality of service (QoS) It is intended to support varied services, such as video, audio, image, and packet data.Therefore we can say that ATM is the culmination of all the development in switching and transmission of data in the last twenty years.

ATM provides an optimum format or protocol family for data , voice and image communications, where cells of each can be intermixed as shown in figure 1.b.

It is more of a compromise. The ATM cells can be transported on different popular digital formats ( e.g. SONET, SDH, E1/T1 etc.).

Figure 1.b. ATM Packets(Cells)

In the term ATM the Asynchronous refers to the manner in which bandwidth is allocated among connections and users. Bandwidth is divided into time slots of fixed length. These time slots are allocated for user information as needed and therefore do not have predetermined temporal positions. Instead of identification of the connection by temporal position, the time slots are identified with explicit prefix label. Transfer mode is term indented to signify that it is multiplexing and switching technique.

 

ATM was born out of standardization efforts for broadband ISDN which began in the ITU-T (formerly CCITT - The International Telegraph and Telephone Consultant Committee ) in the mid 1980’s. Broadband ISDN which is also known as ISDN-B was created conceptually as an extension of ISDN so it functions as a communication network that can provide integrated broadband service such as high speed data service, video phone, CATV services along with the traditional ISDN services such as normal phone and telex .

3.1. ISDN-B & ATM

The Integrated Services Digital Network ( ISDN ) carried a basic channel that could operate at 64kbps ( B-channel) and combinations of this and others ( D-channels ) formed the basis of communication on the network.

However, at the same time, the demand for high-speed packet communication and video communication ( developments in gigabit networks - ATM ) increased led to the need for high-speed and broadband services.

Broadband-ISDN was created which conceptually is just an extension of ISDN so it functions as a communication network that can provide integrated broadband services. This diversity of service meant transmission speeds in the region of 155Mbps, 622Mbps and 2.4Gbps, and hence a means of transmitting and switching at these speeds. While SDH was used for transmission, cell relay emerged as the solution to the switching problem and than combining them together resulted in ISDN-B. The only problem in ISDN-B was to handle switching for broadband where signals range is from 10s of bps and 100s of Mbps and service time ranges from few seconds to several hours.

All the efforts to make the ISDN-B a reality, finally gave birth to ATM , this allowed ISDN-B to came into existence and handle switching and signaling problems.

3.2. ITU-T Recommendations

Basic principals of ATM as put forward in ITU-T Recommendation I.150 are : Ref. ⁴ CCITT Red Book.

• ATM is considered as a specific packet oriented transfer mode based on fixed length cells. Each cell consists of an information field and a header, which is mainly used to determine the virtual channel and to perform the appropriate routing. Cell sequence integrity is preserved per virtual channel.
  
  
• ATM is connection-oriented. The header values are assigned to each section of a connection for the complete duration of the connection. Signaling and user information are carried on separate virtual channels.
  
  
• The information field of ATM cells is carried transparently through the network. No processing like error control is performed on it inside the network.
  
  
• All services (voice, video, data ) can be transported via ATM, including connection less services. To accommodate various services an adaptation function is provided to fit information of all services into ATM cells and to provide service specific functions (e.g. clock recovery, cell loss recovery etc.).
• 
  

• 
  
  4.1. ATM Interfaces & Networking
  An ATM network may be considered a set of ATM switches interconnecting by point-to-point ATM links or interface. ATM switches support two kind of interfaces:

     -- UNI (User network interface)
     -- NNI (Network node interface)

  UNI connect ATM end systems such as hosts, routers to an ATM switch and NNI connects two ATM switches together. An example of ATM network interface is shown in Figure 4.1 .The access rate of UNI is 155.52 Mbps (symmetric in both direction) or 622.08. At 155.52 Mbps access the cell transport bit rate is 149.76 Mbps (Ref. 5 pp254). The actual rate of exchange is determined before the connection establishment between both ends and remainder of transmission capacity is filled with idle cells (Figure 1.b) .The access rate for NNI is also same as UNI. An example of network model is shown in figure 4.a.
  
  

  Figure : 4.a Example of an ATM Network

  In addition ATM network supports relevant call control and network management function. General networking aspects such as virtual path & channels, traffic control and ATM-layer management are discussed in the further sections.

  4.2. ATM Packet

  The ATM packet or cell comprises of 53bytes (as specified by T1S1 sub-committee). Five of bytes makes up the header field and remaining 48 bytes form user information field. The 48bytes of payload may optionally contain a 4 byte ATM adaptation layer and 44 bytes of actual data, or all 48 bytes may contain data based on bit in the control field of the header.UNI and NNI packet format is shown in figure 4.2.a and 4.2.b

  The associated bit sizes for UNI and NNI are shown in Table 4.2.c .For further detail of the header fields (GFC , VPI , VCI , PT, CLP and HEC) can be found in the glossary.

  Functions	UNI	NNI
  Generic Flow Control	4	0
  Virtual Path ID	8	12
  Virtual Channel ID	16	16
  Payload Type	3	3
  Cell Loss Priority	1	1
  Header Error Check Field	8	8

  4.3. ATM layering

  

  4.3.1. ATM adaptation layer

  The purpose of AAL is to isolate the higher layers from the specific characteristics of the ATM layer by mapping the higher layer protocol data units (PDUs) into the information field of the ATM cell and vice versa.

  4.3.2. ATM layer

  The ATM layer is completely independent from the physical medium. The layer generates and extracts the header of the ATM cell, this function of the layer is called encapsulation. This function can also include the translation from receive service point identifier (SAP) to virtual path (VP) and virtual circuit (VC) identifier. In a switch the ATM layer handles the incoming calls. The ATM layer also responsible for traffic management functions.

  4.3.3. Physical Layer

  The Physical layer purpose is to collect and organize ATM cells sent down from the ATM layer, transport them to the physical medium (e.g. fiber optic cable) and also perform the reverse of the process.

  4.4. ATM Traffic

  4.4.1. ATM Traffic Control

  Once the connection is established the network needs certain traffic control capabilities to deal with bandwidth allocation and to provide the an efficient network which copes with potential error within the network.
  
  Another prominent feature of ATM is Traffic shipping feature, this is used to prevent excessive back-to-back cells being generated during the peak transmission of VBR data sources which can effect the ability of other ATM end-station to obtain access.

  4.4.2. Combined bandwidth

  ATM combines the salability of circuit switching with the packet switching. ATM sets the connections before any traffic exchange. It reserve the required bandwidth and specify the connection type (CBR -Constant Bit Rate / VBR- Variable Bit Rate).Other types of traffic such as UBR (Unspecified Bit Rate) and ABR (Available Bit Rate) can also be handled. ABR is reserved bandwidth service and the bandwidth is reserved as it is available instantaneously .If all the bandwidth is reserved , no other service will be allowed within the medium. Figure 4.4.2.a represents the understanding of the above statements .

  

   

  Figure : 4.4.2.a

  The difference between ATM and shared-medium technologies is that shared-medium technologies such as Ethernet devices must share bandwidth between them but on ATM connection each device can have as much of the bandwidth as it needs and much duration its needs it for.

• 
  
  The Telecommunication standardization sector (ITU-T, known as the CCITT until march 1993) of International Telecommunication Union (ITU), and in particular ITU-T Study Group 13 ( formally, CCITT study group XVIII) has been responsible for ISDN-B since 1985.The first milestone was the year 1988 Recommendation I.21, which established ATM as the target transfer mode for ISDN-B and provides the guidelines for subsequent ATM standards. Major 1990-92 recommendations can be seen in the table available in appendix-B (Ref. 3 pp23-24)

  In Europe , the standards are being supported and investigated by ETSI and in United States T1S1 subcommittee (ANSI sponsored) is responsible for ATM support and investigation.

  Most of the key specifications and standards for private ATM networks are being developed at the ATM Forum and the Internet Engineering Task Force (IETF). The former is strictly an "implementers agreement" body, clarifying the use of standards developed at other ATM standards bodies, such as the ITU-T and the ANSI T1S1 Committee.

  In practice, the ATM Forum has considerable extended such standards for private network specific requirements, and has created entirely new specifications, such as LAN Emulation and the P-NNI protocols. The ATM Forum specifications can hence be considered the de facto standards for private network ATM deployment. The specification forwarded by the ATM fourm and ITEF can be seen in the Appendix- (Ref. 9 pp57-58)

• 
  
  6.1. Discussion

  6.1.1. Future Challenges

  This report has explained the working and significant of ATM networks but there is still a great need of improvement in certain areas. Statistical multiplexing in an ATM network is one of important issue. Telecommunications companies in the Europe, US and Japan as well as research organizations and standards committees are actively investigating, how to do improve statistical multiplexing so that the link bandwidth in an ATM network is utilized efficiently, and the quality of service requirements of delay and loss for different types of real time and non real time as well as bursty and continuous traffics are also satisfied during periods of congestion.

  The development in ATM will not effect the common user as the user basically gets access to ATM network through a well structured interface "User Network Interface" (UNI) and network does the rest of job and ensures the qualitative connection .On the other side the service providers and ATM vendors like the telecommunications companies are taking responsibilities of ATM development issues in order to provide better and good quality service.

  6.2. Conclusion

  ATM networks has made fast and qualitative communication possible. The switching and multiplexing techniques that integrate different types of traffic (data, voice, video) at the switching level seems the right approach to step into the future telecommunication world.

  6.2.1. Fixed length Packet

  ATM uses fixed length frames or cells. The fixed length cells give several advantages in order to support simultaneous multimedia services (voice, video and data), some of these advantages are as follows:

  • Provides low-latency
  • Simple switching
  • Avoids usage of Echo cancellor

  6.2.1.1. Low-latency

  Large packet size gives rise to the time taken by the switch to process a packet which makes the delay unpredictable introducing variable-latency and make it difficult to provide near real time service (Voice & video need low latency).

  6.2.1.2. Simple switching

  Fixed cell length make the cell Switching possible instead of using conventional routing technologies. The cell switching can be implemented in hardware which make it efficient and cost effective.

  6.2.1.3. Echo cancellor

  ITU-T specification states that a maximum delay of 24ms is permissible for echo on telephony circuits and if the delay is longer than echo-canceller must be used (increases the cost). The larger cell increases the packet station delay. A cell of 48 bytes will take around 6ms to fill, and 5ms to transmit over the distance of 1000Km using fiber optic medium which means that the round trip delay is approximately 22ms (with in ITU-T limits.).

  6.2.2. Technical & Commercial Point of View

  From all the facts stated in the above sections of the report it can be said that ATM has provided a edge to the technology and a great deal of benefits to the network operators as well as the service provides. ATM offers many advantages from both technical and commercial point of view , out of some are as follows:-

  • Switching to different service with out changing underlying network.
  • It is simple and easy to administrated than multiple sub networks
  • Bandwidth is allocated dynamically which make it possible to achieve efficiency and QoS.
  • There is no need for peak-rate allocation.
  • Potential for granular and flexible control of network traffic.

  By implementing ATM multi-service networks , carriers and service provides can :- (Ref. 6)

  • Construct a network at optimum cost
  • Offer end-to-end broadband service and eliminate bottle necks
  • Profit from the salability by expanding from a small configuration as the need arises.
  • Consolidation of the networks and saving of numerous network elements.
  • Step into the Internet business as an interment service provider (ISP) or offer Internet service to companies.
  • provide business solution with virtual private network services.
  • offer services such as data storage and backup for small business.

  Despite of the above mentioned advantages there are some disadvantages. The asynchronous nature of ATM network can result in cell loss and variable cell delays in the network. Another challenging issue is the feature of supporting the multiple traffic classes with different quality of service requirement, while maximizing statical sharing and utilization of network resources.

7.1. Books

1. Hac, A. Multula, " Synchronous Optical Network and broadband ISDN Protocol", Computer, Nov,1989, pp. 26-34
2. Roger L. Freeman, "Telecommunication System Engineering 3^rd Ed.", John Willey & Sons , Inc. (Chapter 15)
3. Thomas M. Chen, Stephen S. Lui , " ATM Switching Systems" 1995 Artch House, Boston London.
4. CCITT, "Red Book" Volume I-IV
5. S.Ramabhadran, "Communication Engineering", 1993, Satya Prakashan,Tech India Publications. pp 243-315

7.2. Journals & Reports

1. Siemens Telecommunications, "SIEMENS Telecom Report,", Volume 20 (Available in University of Brighton Library)
2. K-NET , "A Guide to ATM",K-NET Ltd,Hampshire,RG29 1AB,UK (Ph: 01256-709200)
3. K-NET , " ATM News", Volume 4.No.3 November 1997
4. Antony Alles, " ATM Internetworking", May 1995. Cisco Systems, Inc. (e-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it )

7.3. World Wide Web

1. URL: http://www.comsworld.com - ATM links
2. URL: http://www.cisco.com - A Report on ATM (Ref.9)
3. URL: http:// www.atm-fourm.com - ATM Fourm Page
4. URL: http:// www.k-net.co.uk - ATM Providers Page
5. URL: http://dxcoms.cern.ch - ATM Specifications

Bit-Timing Information:

Conversion of data bit flow into a waveform adapted to a physical medium, or the reverse conversion process, insertion or extraction of timing information and line coding and decoding. (Ref. 8)

Generic Flow Control (GFC):

Although the primary function of this header is the physical access control, it is often used to reduce cell jitters in CBR services, assign fair capacity for VBR services, and to control traffic for VBR flows. Such functionality requires the power to control any UNI structure, be it a ring, a star, some bus configuration, or any combination of these.

Physical medium:

This depends on the transmission medium - if, for example, optical fibres are used, then the function will be related to that particular medium.

Payload Type (PT) / Cell Loss Priority (CLP) / Header Error Control (HEC):

When user information is present or the ATM cell has suffered traffic congestion then the PT field will yield this information. The CLP bit is used to tell the system whether the corresponding byte is to be discarded during network congestion . ATM cells with CLP=0 have a priority in regard to cell loss than ATM cells with CLP=1. Therefore, during resource congestions, CLP=1 cells are dropped before any CLP=0 cell is dropped. HEC is a CRC byte for the cell header field and is used for sensing and correcting cell errors and in delineating the cell header

Virtual Path Identifier / Virtual Channel Identifier (VPI/VCI):

The role of the VPI/VCI fields is to indicate Virtual Path or Virtual Channel identification numbers, so that the cells belonging to the same connection can be distinguished. A unique and separate VPI/VCI identifier is assigned in advance to indicate which type of cell is following, unassigned cells, physical layer OAM cells, meta-signaling channel or a generic broadcast signaling channel

 

10.1. ATM Implementations ( Case Studies )

10.1.1. Euro-tunnel (Ref. 8, K-NET News Bulletin pp 1)

Eurotunnel has decided to install a high speed ATM backbone. This is to be provided by K-NET at a cost of £½ million. The purpose of installing a high-speed ATM backbone network is to allow Eurotunnel to integrate its day-to-day administration voice and data services between Folkstone and Calais. The network will allow Eurotunnel to provide increased levels of service to telephone sales, and booking and billing services as well as improving voice handling and enabling faster data services. Eurotunnel’s system uses ASX-1000 ATM switches and ES-3810s Ethernet switches provided by FORE systems. The system is designed to make it as future-proof as possible, hence the use of ATM technology.

10.1.2.Microsoft (Ref. 8, K-NET News Bulletin, pp 2-3)

Microsoft’s selection of ATM has helped popularise the view that ATM is the current backbone of choice. Microsoft’s endorsement is two-fold, firstly it has selected a range of ATM products supplied by FORE systems to upgrade the backbone of its Redmond site in Washington

Secondly, packages are currently being developed to allow Microsoft’s operating systems to detect and fully utilise any available ATM connections. These facts are likely to have a major effect on the market

10.1.3. LIMIT (Ref. 7, K-NET ATM Step-by-Step Guide pp27)

LIMIT (Leeds Institute for Minimally Invasive Therapy) was formed in 1994. Its purpose is to facilitate the training of surgeons in the use of minimally invasive medical procedures. LIMIT required a system to transmit high-quality video with minimal latency and two-way audio between its operating theatres (based at Leeds General Infirmary) and its lecture theatres at St. James’ Hospital (8km away).

The Solution adopted was to establish a metropolitan area network providing a 155Mbit/s link between the two sites. To achieve this, a company called K-NET was asked to provide the equipment. The equipment supplied was a set of FORE systems ATM switches. The system makes use of K-NET’s Cellstack video codec to set QOS parameters at such levels as to guarantee the availability of enough bandwidth to deliver near-broadcast quality images. Thanks to this system, LIMIT is now able to provide faster training and thus reduce the burden of training on a single institution. The system is likely to expand once other institutions are suitably equipped.

10.1.4. ORACLE (Ref. 7, K-NET ATM Step-by-Step Guide pp27)

Since 1994, the Oracle Corporation has been expanding its UK operations. The company has also moved its headquarters and data center operations to Reading. This expansion and relocation program led to the need for improved inter-site data and voice traffic capabilities. The company realised that its existing FDDI technology deployed in the LAN would not be sufficiently scaleable to meet their new requirements. The company realised that ATM could provide far better salability to meet the needs of the project. Once fiber links became available in their area, Oracle selected K-NET to install FORE systems’ ATM set up. Oracle’s Campus WAN was believed to be the first of its kind in Europe

11.1. ITU-T 1990-92 Recommendations Table

Table: 11.a Major 1990-92 ITU-T Recommendations(Ref. 3, pp 24)

No	Description
I.121	Broadband aspects of ISDN-B
I.150	ISDN-B ATM functional Characteristics
I.211	ISDN-Service aspects
I.311	ISDN-B General network aspects
I.321	ISDN-B protocol reference model and its applications
I.327	ISDN-B functional architecture
I.361	ISDN-B ATM layer specification
I.362	ISDN-B ATM adaptation layer functional description
I.363	ISDN-B ATM adaptation layer specification
I.371	Traffic control and congestion control in ISDN-B
I.413	ISDN- user network interface
I.432	ISDN-B user network interface physical layer specification
I.610	ISDN-B operation and maintenance principals and functions

11.2. STANDARDS AND SPECIFICATIONS (ITEC & ATM Forum)

B.1 Completed Specifications—ATM Forum

1. UNI 3.0 Contents: Physical layer, ATM layer, OAM cell operation, ILMI, UNI signaling.
2. UNI 3.1 Contents: Bug fixes to UNI 3.0, alignment with completed ITU-T SSCOP and signaling standards.
3. LANE Phase 1 Contents: LUNI protocol
4. IISP Contents: UNI 3.0/3.1 based static routing NNI protocol

B.2 Completed Specifications—IETF

1. RFC 1483 Contents: Multi-protocol Encapsulation
2. RFC 1577 Contents: Classical IP Over ATM protocol
3. RFC 1626 Contents: Default MTU for Classical IP
4. RFC 1755 Contents: Signaling guidelines for Classical IP

B.3 Pending Specifications—ATM Forum

P-NNI Phase 1 Contents: QoS based NNI routing, hierarchical network model. Expected Completion Date: Q3 1995

1. ABR Congestion Control Contents: Best effort traffic class and rate based con-gestion control mechanism. Expected Completion Date: Q3/Q4 1995
2. UNI 4.0 Signaling Contents: ABR signaling, leaf initiated joins, QoS nego-tiation,VP signaling, proxy signaling etc. Expected Com-pletion Date: Q3/Q4 1995
3. MPOA Contents: Multiprotocol transport over ATM. Expected Completion Date: Q1/Q2 1996
4. LANE Phase 2 Contents: L-NNI specification for redundant servers. Expected Completion Date: Q1/Q2 1996

B.4 Pending Specifications—IETF

1. NHRP Contents: Cut through routing extensions to Classical IP model. Expected Completion Date: Q2/Q3 1995
2. Multicast Support in 1577 Contents: Multicast registration services in Classical IP. Expected Completion Date: Q2/Q3 1995
3. IPv6 (IPng) Contents: Family of specifications for complete IPv6 protocol. Expected Completion Date: Q4 1995
4. RSVP Contents: Resource reservation protocol for IP. Expected Completion Date: Q3 1995
5. PIM Contents: Protocol independent multicast protocol for IP. Expected Completion Date: Q3 1995

6. 

   A SURVEY OF ATM TRAFFICMANAGEMENT By CISCO Systems, Inc.
   (Ref. 9, pp 54)

   One of the primary benefits of ATM networks is that they can provide users with a guaranteed Quality of Service (QoS). To do this, the user must inform the network, upon connection set-up, of both the expected nature of the traffic that will be sent along the connection, and of the type of quality of service that the connection requires. The former is described by a set of traffic parameters, while the latter is specified by a set of desired QoS parameters. The source node must inform the network of the traffic parameters and desired QoS for ach direction of the requested connection upon initial set-up; these parameters may be different, however, in each direction of the connection.ATM networks offer a specific set of service classes, and at connection set-up, the user must request a specific service class from the network for that connection. Service classes are used by ATM networks to differentiate between specific types of connections, each with a particular mix of traffic and QoS parameters, since such traffic may need to be differen-tiated within the network, for instance, by using priorities to allow for the requested behavior. The current set of QoS classes 63 , which the Forum is defining for UNI 4.0 is as follows:

7. Continuous Bit Rate [CBR]: End systems would use CBR connection types to carry constant bit rate traffic with a fixed timing relationship between data samples, typically for circuit emulation.
8. Variable Bit Rate—Real Time [VBR(RT)]: The VBR(RT) service class is used for connections that carry variable bit rate traffic, in which there is a fixed timing relationship between samples; for instance, for such applications as variable bit rate video compression.
9. Variable Bit Rate—Non-Real Time 64 [VBR(NRT)]: The VBR(NRT) service class is used for connections that carry variable bit rate traffic in which there is no timing rela-tionship between data samples, but a guarantee of QoS (on bandwidth or latency) is still required. Such a service class might be used for Frame Relay internetworking, in which the Committed Information Rate (CIR) of the Frame Relay connection is mapped into a bandwidth guarantee within the ATM network.
10. Available Bit Rate [ABR]: The ATM Forum is currently focusing its work on the ABR service ([Forum9], [Jain], [Hughes]). As with the VBR(NRT) service, ABR supportsvariable rate data transmissions and does not preserve any timing relationships between source and destination. Unlike the VBR(NRT) service, however, the ABR service does not provide any guaranteed bandwidth to the user. Rather, the network provides a "best effort" service, in which feedback (flow control mechanisms) is used to increase the bandwidth available to the user—the Allowed Cell Rate (ACR)—if the network is not congested and to reduce the bandwidth when there is congestion. Through such flow control mechanisms, the network can control the amount of traffic that it allows into the network, and minimize cell loss within the network due to con-gestion. The ATM Forum is currently working on a "rate based" mechanism for ABR congestion control, where Resource Management (RM) Cells or the explicit forward con-gestion indication (EFCI) bit within ATM cells are used to indicate the presence of ongestion within the network to the source system. A specified traffic pacing algorithm, controlling the ACR, is used at the source to control the traffic rate into the network, based either upon the number of RM cells received with a congestion indication or an explicit rate indication from the network. Refer to [Forum9] for more details.ABR is designed to map to existing LAN protocols that opportunistically use as much bandwidth as is available from the network, but can either back off, or be buffered in the presence of congestion. ABR is hence ideal for carrying LAN traffic (for instance, using LAN mulation) across ATM networks. The ABR service can optionally provide a uaranteedMinimum Cell Rate (MCR) for an ABR connection, but the exact nature of this guarantee is currently a matter of debate within the ATM Forum.
11. Unspecified Bit Rate [UBR]: The UBR service does not offer any service guarantees. The user is free to send any amount of data up to a specified maximum while the network makes no guarantees at all on the cell loss rate, delay, or delay variation that might be experienced. The UBR service is currently the best match to LAN protocols, given that the ABR specification has yet to be completed. As of the time of writing, it ppeared that the ABR speci-fication would not be completed until well into the second half of 1995. Deployment of BR compliant equipment will likely take even longer. In the meantime, UBR is the only service currently vailable for data transport. Since UBR does not have any flow control mechanisms, however, to control or limit congestion, it will be important that ATM switches either implement pre-standard congestion control-mechanisms.

12. 13.1 List of Acronames

    A

    AAL ATM Adaptation Layer
    ABR Available Bit rate
    ATD Asynchronous Time Division
    ATM Asynchronous Transfer Mode

    B

    BIP Bit Interveald Parity
    B-ISDN Broadband Integrated Services Digital Network

    C

    CAC Connection Admission Control
    CBR Constant Bit Rate
    CCITT Comite Consultatif Internationale Telegrapgique Et
    CLP Cell Loss Parity
    CRC Cycle Redundancy Check

    D

    DDC Data Communication Channel

    E

    EOC Embedded Operation Channel
    ESS Electronic Stitching System
    ETSI European Telecommunication Standards Institution

    F

    FDDI Fiber Distributed Data Interface
    FDM Frequency Division Multiplexing

    G

    Gbps Giga Byte Per Seconds
    GFC Generic Flow Control

    H

    HEC Header Error Control

    I

    IDN Integrated Digital Network
    IN Intelligent Network
    ITU International Telecommunication Union
    ITU-T I TU Telecommunication Standardization Sector

    L

    LAN Local Area Network
    LOF loss of frame

    M

    MIN Multistage interconnection Network

    N

    NDF New Data Flag
    NNI Network Node Interface

    O

    OSI Open System Interconnection

    P

    PDU Protocol Data Unit
    PK Physical Medium
    PT Payload Type

    Q

    Qos Quality of Service

    R

    RDI Remote Defect Indication

    S

    SM System Management
    SONET Synchronous Optical Network
    SSP Service Switching point
    STP Signal Transfer Point

    T

    TCP Transmission Control Protocol
    TDM Time Division Multiplexing

    U

    UME UNI Management Entry
    UNI User Network Interface

    V

    VBR Variable Bit Rate
    VC Virtual Channel
    VCC Virtual Channel Connection
    VP Virtual Path
    VPC Virtual Path Connection
    VPI Virtual Path Identifier

    W

    WAN Wide Area Network