Multicast Papers Appeared in 1992

S. Armstrong, A. Freier, and K. Marzullo, "Multicast transport protocol," Request for Comments (Informational) RFC 1301, Internet Engineering Task Force, Feb. 1992.

Abstract: This memo describes a protocol for reliable transport that utilizes the multicast capability of applicable lower layer networking architectures. The transport definition permits an arbitrary number of transport providers to perform realtime collaborations without requiring networking clients (aka, applications) to possess detailed knowledge of the population or geographical dispersion of the participating members. It is not network architectural specific, but does implicitly require some form of multicasting (or broadcasting) at the data link level, as well as some means of communicating that capability up through the layers to the transport. Keywords: reliable transport, multicast, broadcast, collaboration, networking.

Anthony Alles and Keith McCloghrie, "Unified ATM addressing proposal," Dec. 1992.

Abstract: This contribution proposes a unified addressing scheme for use with both private and public ATM networks. It builds upon the requirements for addressing adapted at the November meeting and proposes a small, simple set of structured addresses, which will facilitate the routing of ATM signaling messages. It also incorporates support for structured multicasting addresses, which are essential for the support of ATM multicast groups in the wide area.
Keywords: ATM; addressing

Anonymous, "Xpress transfer protocol (XTP) specification version 3.6," tech. rep., Protocol Engines, 1992 ??

Abstract: This document defines XTP message formats and peer-to-peer message exchanges. It assumes the reader is familiar with network protocols such as Delta-T, NETBLT, and IP for the motivation, philosophy, and explainations that are largely omitted here. It is also assumed that the reader is familiar with the C programming language. Synopsis. SECTION 1 of this document consists of background and introductory material. Except for the glossary of terms it is separate from the protocol specification which is contained in sections 2 and 3. SECTION 2 defines the syntax of XTP - i.e. the packet formats, addressing conventions and encapsulation formats - and also defines the meaning and interpretation of control fields and state elements in protocol messages. In some cases the definitions in section 2 include procedures for generating or interpreting specific fields. Tables and figures that define the syntax of XTP packets are collected together in Appendix E for guick reference. SECTION 3 defines the end-to-end protocol procedures. The description is modular and reflects the structure of the design. The major sections are management procedures, data flow procedures, multicast procedures, and timers. Subsections within each of these major sections define the difference protocol mechanisms such as flow control, rate control and error control. Several appendices that further explain or enhance sections 2 and 3 are attached to the document. These include a description of the checksum function, detailed description of multicast heuristics, state tables for connection management, a sample application interface, and the summary of packet structures.
Keywords: XTP; Xpress Transfer Protocol; network protocol; design network protocol analysis; multicast

Rosario Aiello, Elena Pagani, and Gian Paolo Rossi, "Causal ordering in reliable group communications," in SIGCOMM Symposium on Communications Architectures and Protocols (Deepinder P. Sidhu, ed.), (San Francisco, California), pp. 106-115, ACM, Sept. 1993. also in \em Computer Communication Review 23 (4), Oct. 1992.

Abstract: In this paper we present a solution to the causal reliable multicast problem. User processes generate separate sequences of messages and specify the causal relation among them according to some application need; the algorithm ensures that the messages within the same sequence are delivered to all active, i.e., both correct and faulty, processes in the group, or to none of them, and are processed according to their causal order. Messages belonging to different sequences can be concurrently processed. This problem has few solutions presented in the literature; in common with a part of them, the algorithm we describe has the centralized approach and the use of history buffers to recover from omission failures. The differences mainly concern the mechanism we devised to recover from crash failures, that avoids resorting to specialized protocols. As a consequence, under failure conditions, the algorithm performs better than other proposals in terms of both network load and throughput without affecting the performances under reliable conditions. Further, it allows to implement the most general interpretation of causality and it does not require any particular service to the underlying transport protocol.
Keywords: multicast; causality; reliable multicast; group communications; distributed algorithms

Tony Ballardie, Paul Francis, and Jon Crowcroft, "Core based trees (CBT)," in SIGCOMM Symposium on Communications Architectures and Protocols (Deepinder P. Sidhu, ed.), (San Francisco, California), pp. 85-95, ACM, Sept. 1993. also in \em Computer Communication Review 23 (4), Oct. 1992.

Abstract: One of the central problems in one-to-many wide-area communications is forming the delivery tree - the collection of nodes and links that a multicast packet traverses. Significant problems remain to be solved in the area of multicast tree formation, the problem of scaling being paramount among these. In this paper we show how the current IP multicast architecture scales poorly (by scale poorly, we mean consume too much memory, bandwidth, or too many processing resources), and subsequently present a multicast protocol based on a new scalable architecture that is low-cost, relatively simple, and efficient. We also show how this architecture is decoupled from (though dependent on) unicast routing, and is therefore easy to install in an internet that comprises multiple heterogeneous unicast routing algorithms.
Keywords: routing; multicast; core-based trees; spanning trees; IP; Internet

R. P. Bianchini and H. S. Kim, "Design of a Nonblocking Shared-Memory Copy Network for ATM," in Proceedings of the Conference on Computer Communications (IEEE Infocom), (Florence), pp. 876-885, 1992.

Keywords: ATM; switching system; multicast; architecture; central buffer; performance evaluation; simulation

Anthony Ballardie, Paul Tsuchiya, and Jon Crowcroft, "Core Based Trees (CBT)," Internet Draft I-D, University College London, Nov. 1992. work in progress.

Abstract: Multicasting is a technique used which allows for group communications either locally or on a wide-area scale. Most local-area networks such as Ethernet and Tokenring provide a multicast service, which has since been exploited by many applications and distributed systems. More recently, multicast capability has been extended to the internetwork using a combination of a distance-vector routing algorithm, a host-to-router group-membership reporting protocol, and the computation of sender-based multicast trees. This, and other approaches to internet multicasting, are not scalable to large groups, especially so for large numbers of groups. This document provides a proposal for a new multicast routing algorithm which provides multicast capability in a datagram internetwork. It is scalable, low-cost and efficient - properties lacking in current internetwork multicast routing protocols.
Keywords: multicast; routing

Stephen Casner and Stephen Deering, "First IETF Internet audiocast," ACM Computer Communication Review, vol. 22, pp. 92-97, July 1992.

Abstract: Describes the IETF audiocast from San Diego (April 1992), including the network multicast backbone, the audio packet format and applications.
Keywords: packet voice; audio; audiocast; IETF; Internet
Annotation: cites Nevot, vat

Andreas Cramer, Manny Farber, Brian McKellar, and Ralf Steinmetz, "Experiences with the Heidelberg multimedia communication system: multicast, rate enforcement and performance," in 4th IFIP Conference on High Performance Networking, (Liège, Belgium), pp. D4-1 - D4-20, IFIP, Dec. 1992.

Abstract: The Heidelberge multimedia communication system, designed at the IBM European Networking Center at Heidelberg, is intended for high-speed data communications and multimedia data exchange. A first version was demonstrated at CeBIT 92. Challenges encountered during the development included multicast on token ring LAN and rate enforcement. Performance of the prototype is also presented in this paper.
Keywords: multimedia; multicast; rate enforcement; policing; token ring; rate control; local area networks

Xing Chen and Jeremiah F Hayes, "Call Scheduling in Multicasting Packet Switching," in Conference Record of the International Conference on Communications (ICC), pp. 895-899, 332.3.1-332.3.5, June 1992.

Abstract: The subject of the paper is the implementation and performance of multicast packet switching in a broadband network environment. In terms of scheduling the transmission of the copies of the packet onto output ports, two basic service disciplines have been defined, 1) one-shot scheduling - all the copies transmitted in the same time slot and 2) call splitting - transmission over several time slots. As subcategories of call splitting, SS specifies that each packet can send at most one copy to the destination per time slot; WS does not carry this restriction. We propose a novel scheme, revision scheduling, to mitigate the HOL blocking effect by sequentially combining the one-shot scheduling and the call splitting disciplines. This paper introduces schematic structures for each category of scheduling, in the form of combinational logic circuits which are designed to resolve the output contentions corresponding to call scheduling disciplines, by incorporating a cyclic priority access policy. In order to compare the performance of various techniques, simulation studies of performance were carried out.
Keywords: multicast; switch design; call scheduling; performance analysis; simulation; ATM

Yee Hsiang Chang, "Remote Conferencing Architecture," research note, MCNC Center for Communications, ??, July 1992.

Abstract: Two important issues should be resolved before building remote conference architecture - network resource management and wide-area multicast address administration. In this paper, we will not discuss the details of these issues, which are still being investigated. Instead, we will point out how these issues are dependent on the session-layer function. This dependency requires one important feature: the connection/configuration set-up with the networks must handle the services of resource management and multicast addressing.
Keywords: teleconferencing; session control

Fabio M. Chiussi and Fouad A. Tobagi, "A Hybrid Shared-Memory/Space-Division Architecture for Large Fast Packet Switches," in Conference Record of the International Conference on Communications (ICC), pp. 905-911, 332.5.1-332.5.7, June 1992.

Abstract: In fast packet switches, regardless of the architecture, implementation constraints are the primary limitations on size and functionality. To accommodate large sizes, a solution has been to interconnect smaller identical modules in a multistage configuration. In this paper, we explore the use of components of different architectural types to build a large fast packet switch. We introduce a hybrid architecture, referred to as the Memory/Space-division/Memory (MSM) switching fabric, which integrates shared memory and space-division switching techniques and makes it possible to build an efficient self-routing switch capable of serving several hundred users. This architecture presents a simple structure, achieves high throughput under a wide variety of traffic conditions, and is easily expandable to support additional functionalities, such as burst switching, hybrid circuit/packet switching operation, and multicasting.
Keywords: ATM switch design switch architecture switch, banyan, Tandem Banyan Switching Fabric (TBSF) Memory/Space/Memory (MSM) banyan, chip design CLP (cell loss probability)

André Danthine, "Esprit Project OSI 95," Computer Networks and ISDN Systems, vol. 25, pp. 384-399, Dec. 1992.

Abstract: The last ten years have seen tremendous change in the communication environment. The changes in network performance have been at the origin of new protocols and of more efficient methods of implementation. The changes in network service capability have raised the issue of offering, at the transport level, services available at the subnetwork level, such as multicast service, synchronous service and broadband service. The changes in application requirements have introduced new requirements for communication. The client/server paradigm is looking for low latency, multimedia for low jitter and many applications are requesting bandwidth. The paper discusses the need for new OSI standards for the transport service and for the transport protocol. The work done in the ESPRIT II project OSI 95 as well as the ISO activities are presented.
Keywords: transport service; transport protocol; LOTOS; multimedia; broadband; high-speed networks; OSI 95; TP4; TCP; congestion control; client/server; request/response; quality-of-service

André Danthine, Yves Baguette, Guy Leduc, and Luc Léonard, "The OSI 95 connection-mode transport service - the enhanced QOS," in 4th IFIP Conference on High Performance Networking, (Liège, Belgium), pp. E2-1 - E2-18, IFIP, Dec. 1992.

Abstract: During the last ten years, tremendous changes have taken place in the communication environments. First, there has been a continuous increase in network performance that has led, for instance, to increasingly high access data rates available in the lower layers. Furthermore, the changes in the offered network services have raised the issue of providing, at the transport level, services already provided at the subnetwork level, such as multicast or synchronous services for example. With the arrival of new applications, such as multimedia or client/server applications, a widening of the application requirements has also been observed. It is this evolving environment that has been at the origin of the ESPRIT II Project OSI 95. An important task in the framework of this project is the definition of an enhanced transport service taking account of the aforementioned evolutions. The enhanced transport service specified for OSI 95 consists of several types of service. The paper presents the connection-mode transport service. We focus mainly on the following original features of our connection-mode service: a new semantics for QoS parameters and the associated negotiation and re-negotiation.
Keywords: OSI; transport protocols; quality of service

Michael E. Gaddis, Rick Bubenik, and John D. Dehart, "A Call Model for Multipoint Communication in Switched Networks," in Conference Record of the International Conference on Communications (ICC), pp. 609-615, 322.5.1-322.5.7, June 1992.

Abstract: We describe a call model for multipoint communication in switched networks. The model provides network clients with dynamic multipoint, multiconnection communication channels, which we term calls. Clients create, manage and manipulate calls using our Connection Management Access Protocol (CMAP). The call model provides basic interconnection services suitable for local and wide area networks, where more sophisticated services can be layered over this substrate. Multimedia workstations, characterized by video, audio and data transfer capabilities, are becoming increasingly popular. When several multimedia workstations are interconnected over a network, the bandwidth requirement is much higher than that with traditional data-only workstations. Shared media networks, such as Ethernet and FDDI, do not have the required bandwidth to support many of these multimedia applications, especially in cases where several multimedia workstations are connected to the same physical network, and where several workstations are communicating with one another simultaneously. Switched Local Area Networks (LANs) offer the potential to alleviate this network botteneck. With switched LANs, such as those based on the Asynchronous Transfer Mode (ATM) technology, each workstation has a dedicated link (in the case of ATM, at 155 megabits of higher). The workstations are interconnected through one or more packet switches, which allow simultaneous transfers at maximum link rate between several workstations. In addition to solving the immediate problem of supporting multimedia traffic in local areas, switched networks have the potential to scale to Metropolitan Area Network (MAN) and Wide Area Network (WAN) sizes. The ATM standard is being developed currently by CCITT for the public communication WANs. As such, switched ATM LANs will likely be able to interface directly to the public networks, perhaps creating one ubiquitous global network. This paper describes a call model designed for switched ATM networks applicable to both LAN and WAN environments. The model supports multipoint, multiconnection communication channels, which we term calls. Calls are allowed to change dynamically during their lifetime, in terms of the number of participants, the number of connections and the bandwidth of the connections. Our model defines how clients view and interact with the network to create, manage and manipulate calls. It defines the outward functionality of the network, its expected behavior and atomic capabilities. Network clients interact with the network using out Connection Management Access Protocol (CMAP) [1] [11] [12]. The remainder of this paper is organized as follows. In Section 2, we give a brief overview of ATM networks. In Section 3, we describe the ATM standard and the basic data information exchange component, the ATM cell pipe. In Section 4, we describe two types of ATM connections, virtual path and virtual channel, and discuss how they can be used for multipoint communication. In Section 5, we describe the call model. Finally, we cover related work in Section 6 and summarize in Section 7.; references=; [1] R.G. Bubenik, J.D. Dehart and M.E. Gaddis. "Multipoint Connection Management in High Speed Networks." In IEEE Infocom `91; Proceedings of the Tenth Annual Joint Conference of the IEEE Computer and Communications Societies, pages 59-68, April 1991. [4] ANSI T1S1 Technical Sub-Committee. Broadband Aspects of ISDN Baseline Document. T1S1.5/90-001, June 1990. [6] CCITT, Report of Working Party XVIII/8 (General BISDN Aspects), Study Group XVIII-Report R70, July 1991 (Report on Geneva Meeting, 11-28 June 1991. [11] J.D. Dehart, M.E. Gaddis and R.G. Bubenik, "Connection Management and Access Control (CMAP) Specification," Washington University, Department of Computer Science, Technical Report 92-01, January 1992. [12] M.E. Gaddis, R.G. Bubenik, J.D. Dehart. "Connection Management for a Prototype Fast Packet ATM BISDN Network.
Keywords: call control call model multicast multipoint ATM VCI (virtual channel identifier) VPI (virtual path identifier) VP (virtual path) VC (virtual channel) signalling LAN

Van Jacobson, "sd, the LBL session directory," Manual page, Nov. 1992.

Abstract: sd is a multicast-based directory for multimedia conferencing sessions. It allocates multicast addresses from a subspace of class-D IP addresses and periodically multicasts information about the session, such as the media multicast address, port number, encoding, etc.
Keywords: directory; CSCW; conferencing

Xiaofeng Jiang, "Routing broadband multicast streams," Computer Communications, vol. 15, pp. 45-51, January/February 1992.

H. S. Kim and R. P. Bianchini, "Omega network based modular multicast ATM switch," in International Switching Symposium, vol. 2, (Yokohama), p. A7.3, Oct. 1992.

Abstract: In this paper we present the design and implementation of an ATM switch for BISDN traffic using standard ATM cells. We propose a new multicast ATM switch that differs significantly from currently proposed switches in terms of switch performance and hardware complexity. Main characteristics of the switch architecture include: shared-memory queue, distributed control, multicast capability, low hardware cost.
Keywords: ATM; multicast; BISDN; Omega network; network; architecture; simulation

C. K. Kim and T. T. Lee, "Call Scheduling Algorithms in a Multicast Switch," IEEE Transactions on Communications, vol. COM-40, no. 3, pp. 625-635, 1992.

Abstract: In this paper we develop and analyze call scheduling algorithms for a multicast circuit switch. In particular, we examine two general classes of scheduling algorithms: call packing algorithms and call splitting algorithms. Performance improvement bycall packing examined in this paper is shown to be negligible. in contrast, call splitting algorithms can provide significantlylower blocking by reducing the level of output port contention.
Keywords: STM; circuit switching; multicast; switching; algorithm; performance evaluation; analysis; blocking

Vachaspathi Kompella, Joseph C. Pasquale, and George C. Polyzos, "Multicasting for multimedia applications," in Proceedings of the Conference on Computer Communications (IEEE Infocom), vol. 3, (Florence, Italy), pp. 2078-2085 (9A.1), IEEE, May 1992.

Abstract: We investigate multicast routing for high-bandwidth delay-sensitive applications in a point-to-point network as an otpimization problem. We associate an edge cost and an edge delay with each edge in the network. The problem is to construct a tree spanning the destination nodes, such that it has the least cost, and so that the delay on the path from source to each destination is bounded. Since the problem is computationally intractable, we present an efficient approximation algorithm. Experimental results through simulations show that the performance of the heuristic is near optimal.
Keywords: multicast; routing; Steiner tree

T. H. Lee and S. J. Liu, "A Fair High-Speed Copy Network for Multicast Packet Switch," in Proceedings of the Conference on Computer Communications (IEEE Infocom), (Florence), pp. 886-894, 1992.

Keywords: ATM; switching system; multicast; architecture; multistage interconnection network; Banyan network

Erwin Mayer, "An evaluation framework for multicast ordering protocols," in SIGCOMM Symposium on Communications Architectures and Protocols, (Baltimore, Maryland), pp. 177-187, ACM, Aug. 1992. \em Computer Communication Review, Volume 22, Number 4.

Abstract: A new framework for evaluating multicast ordering protocols is presented. It allows to compare solutions to the problem of ordering multicast messages in an identical way at all receiving sites when multiple senders operate concurrently. A new type of delay measure called synchronization delay forms the basis for this framework. It filters out network-dependent factors and accounts precisely for the excess delay that user messages suffer in order to achieve the ordering property. The usefulness of the evaluation framework is demonstrated by applying it to three protocols covered in the literature. As an example their delay behavior is analyzed for low-traffic environments. The evaluation results allows a user to choose the most suitable protocol for his application.
Keywords: multicast; reliable multicast; ordering; distributed systems; synchronization

S. Okamoto, "Modular Expandable Multi-Stage ATM Cross-Connect System architectue for ATM Broadband Networks," IEICE Transactions on Communications, vol. E75-B, no. 3, pp. 207-216, 1992.

Abstract: This paper describes the design of a large capacity ATM cross-connect system that has a multistage network structure which requires only one type of switch module. The capacity of the proposed system can be easily increased without service interruptions. To realize cell sequence integrity, a time stamp is added to the self-routing tag. Required time stamp length and efficient module size are discussed.
Keywords: ATM; switching network; multistage interconnection network; multicast; Banyan network; self routing; architecture; performance evaluation; simulation

Amy Pearl, "System support for integrated desktop video conferencing," Technical Report TR-92-4, Sun Microsystems Laboratories, Mountain View, California, Dec. 1992.

Abstract: Desktop computers are increasingly used as communication devices. Advances in digital media are making the integration of video into desktop computers practical, both technically and economically. At the convergence of these technologies is computer-integrated interactive video conferencing. This paper discusses the requirements of integrated desktop video conferencing on a networked set of multimedia-capable workstations. Among those requirements are the following: (1) Media-intensive parts of applications should be distributed; a multimedia software platform should provide support for this. (2) Audio and video conferencing require network transparent location and reference of people, media devices, and conferences. The name and remote access reference for a conference must be exportable to client applications. (3) All group support applications that provide remote access require security services. True in any network application, this is more important with live communication streams, such as audio and video. (4) Low latency of an audio connection is more important than synchronization of audio with other timecritical communication, such as video conferencing and user gestures in shared interactive applications. (5) To efficiently support multi-person conferences, multicast networking protocols are essential. A research prototype multimedia platform was evaluated, based on these requirements. This paper presents the lessons learned about system requirements for video conferencing that are not obviously required for single-user multimedia. It also discusses the motivation and requirements for extending the platform to support shared applications not previously considered to have constraints related to time.
Keywords: multimedia; operating system; packet audio; packet video

Joseph C. Pasquale, George C. Polyzos, Eric W. Anderson, and Vachaspathi P. Kompella, "The multimedia multicast channel," in Third International Workshop on network and operating system support for digital audio and video, (San Diego, California), pp. 185-196, IEEE Communications Society, Nov. 1992.

Abstract: The multimedia multicast channel is a dissemination oriented communication abstraction providing a service analogous to that of a cable television broadcast channel. A source transmits multimedia information such as video and audio streams onto a channel, a varying number of receivers ``tune in'' to the channel to receive a selected set of the streams. To support heterogeneity, each receiver may tailor the selected streams to meet individual needs through the use of filters. The design encourages a very loose coupling between the source and the receivers, promoting open-loop control for the underlying network protocols.
Keywords: multimedia; API; multicast; dissemination

Bala Rajagopalan, "Reliability and scaling issues in multicast communication," in SIGCOMM Symposium on Communications Architectures and Protocols (Deepinder P. Sidhu, ed.), (Baltimore, Maryland), pp. 188-198, ACM, Aug. 1992. in \em Computer Communication Review 22 (4), Oct. 1992.

Abstract: The efficiency with which multicast communication can take place is largely determined by the network level support for such communication. Two factors contribute to the complexity of supporting current multicast applications: the lack of reliable multicast transport level mechanisms at the network level and the lack of network support for large scale multicast communication. In this paper, we examine the issues pertinent to eliminating these shortcomings. We first show that internet multicasting algorithms based on reverse path forwarding are inherently unreliable and present a source-tree based reliable multicasting scheme. The new scheme makes use of simple inter-gateway protocols and works on top of previously developed distance vector and link state internet routing schemes. Next, to support large scale applications, we present a scheme for partial multicasting and introduce a new network level operation, called gather. The partial multicasting mechanism allows messages to be delivered to subsets of multicast destinations, while the gather operation aids gateways in selectively suppressing redundant messages, thus reducing the message complexity. Using simulations, we investigate the efficacy of our schemes in supporting a sample application based on multicast communication.
Keywords: multicast; network layer; reliability; source tree; distance vector; link state

S. Sato, T. Aramaki, H. Suzuki, and M. Akiyama, "Cell-distribution-routing architecture for ATM switching network," in International Switching Symposium, vol. 1, (Yokohama), p. P17, Oct. 1992.

Abstract: This paper proposes a three-stage cell-distribution-routing ATM switching network which is high performance for various connections such as CBR, VBR, and multicast traffic. This paper first discusses the relationship between cell-level QOS and nonblocking. Further aspects: description of a simple method to prevent cell-sequence disorder, evaluation of the performance of the network using simulation and analysis
Keywords: ATM; switching block; high-performance switching network; bursty traffic; multicast; architecture; nonblocking; simulation; performance evaluation

Eve M. Schooler, "The impact of scaling on a multimedia connection architecture," in Third International Workshop on network and operating system support for digital audio and video, (San Diego, California), pp. 302-307, IEEE Computer and Communications Societies, Nov. 1992.

Abstract: As the last two meetings of the Internet Engineering Task Force (IETF) have shown, Internet teleconferencing has arrived. Packet audio and video have now been multicast to approximately 170 different hosts in 10 countries, and for the November 1992 meeting the number of remote participants is likely to be substantially larger. Yet the network infrastructure to support wide scale packet teleconferencing is not in place. This paper discusses the impact of scaling on our efforts to define a multimedia teleconferencing architecture. Three scaling dimensions of particular interest include: (i) very large numbers of participants per conference, (ii) many simultaneous teleconferences, and (iii) a widely dispersed user population. Here we present a strawman architecture and describe how conference-specific information is captured, then conveyed among end-systems. We provide a comparison of connection models and outline the tradeoffs and requirements that change as we travel along each dimension of scale.
Keywords: multimedia; conference control; scaling

Timothy W. Strayer, Bert J. Dempsey, and Alfred C. Weaver, XTP - the xpress transfer protocol. Addison-Wesley Publishing Company, 1992.

Abstract: Chapter 1 Foundations of the Xpress Transfer Protocol Section 1.1 Why ANY New Protocol? Section 1.1.1 Changes Below Section 1.1.2 Changes Above Section 1.1.3 Changes Within Section 1.1.4 Repair or Replace? Section 1.2 Why THIS New Protocol Section 1.2.1 Functional Enhancements Section 1.1.2 Performance Enhancements Section 1.3 History of XTP Section 1.4 Organization of Book Chapter 2 Network Concepts and the XTP Architectural Model Section 2.1 Network Concepts Section 2.1.1 The OSI Reference Model Section 2.1.2 Data Transfer Models Section 2.1.3 Services and Service Access Section 2.1.4 Transport Layer Section 2.1.5 Network Layer Section 2.2 XTP Architecture Section 2.2.1 XTP Transfer Layer Architecture Section 2.2.2 XTP Data Communication Model Section 2.2.3 XTP Data Delivery Service Requirements Chapter 3 Influential Protocols Section 3.1 Conventional Transport Section 3.1.1 TCP Section 3.1.2 ISO Transport Protocol Section 3.1.3 Conventional Mechanisms Section 3.2 Delta-t Section 3.3 NETBLT Section 3.4 GAM-T-103 Military Real-Time Local Area Network Architecture Section 3.5 Versatile Message Transaction Protocol Section 3.6 Datakit and the Universal Receiver Protocol Section 3.7 Summary Chapter 4 Protocol Procedures Section 4.1 Introduction Section 4.1.1 XTP Packet Structure Overview Section 4.1.2 XTP Packet Formats Section 4.1.3 XTP protocol Procedures Section 4.1.4 Conventional Notations Section 4.2 Association Management Procedures Section 4.2.1 Establishing an Association Section 4.2.2 Maintaining an Association Section 4.2.3 Terminating an Association Section 4.2.4 Interesting Paradigms Section 4.2.5 Multicast Association Section 4.3 Path Management Procedures Section 4.3.1 Path Establishment Section 4.3.2 Path Maintenance Section 4.3.3 Path Release Section 4.4 Data Flow Procedures Section 4.5 Flow Control Procedures Section 4.6 Rate Control Procedures Section 4.7 Error Control Procedures Section 4.7.1 Checksums Section 4.7.2 Acknowledgements and Retransmissions Section 4.7.3 Timers Section 4.7.4 Synchronizing Handshake Section 4.7.5 Time-to-Live Section 4.7.6 Error Notification Chapter 5 Packet Structures Section 5.1 Introduction Section 5.1.1 Notation Convention Section 5.1.2 Segments Section 5.2 XTP Header Section 5.2.1 Route Field (route) Section 5.2.2 Time-to-Live Field (ttl) Section 5.2.3 Command Field (cmd) Section 5.2.4 Key Field (key) Section 5.2.5 Synchronize Field (sync) Section 5.2.6 Sequence Number Field (seq) Section 5.2.7 Delivered Sequence Number Field (dseq) Section 5.2.8 Sort Field (sort) Section 5.2.9 Data Length Field (dlen) Section 5.2.10 Header Checksum Field (hcheck) Section 5.3 XTP Trailer Section 5.4 Control Segment Section 5.4.1 Rate Control Fields (rate and burst) Section 5.4.2 Synchronize Echo Field (echo) Section 5.4.3 Time Synchronization Fields (time and techo) Section 5.4.4 Key Exchange Field (xkey) Section 5.4.5 Route Exchange Field (xroute) Section 5.4.6 Allocation Field (alloc) Section 5.4.7 Received Sequence Number Field (rseq) Section 5.4.8 Selective Retransmission Fields (nspan and spans) Section 5.4.9 The Reserved Fields (rsvd) Section 5.5 Information Segment Section 5.5.1 Data Segment Section 5.5.2 Address Segment Section 5.5.3 Data and Address Segments Section 5.5.4 Management Segment Chapter 6 Packet Formats Section 6.1 FIRST Packet Section 6.2 DATA Packet Section 6.3
Keywords: book network protocol design network protocol analysis XTP multicast

Harry Santoso and Serge Fdida, "Transport layer multicast: an enhancement for XTP bucket error control," in 4th IFIP Conference on High Performance Networking, (Liège, Belgium), pp. G2-1 - G2-17, IFIP, Dec. 1992.

Abstract: Transport layer multicasting is defined as a transfer of a transport protocol data unit from one sender to a well defined set of receivers. Previous work in the area of multicasting is concentrated on the MAC and IP layers. In this paper, we address the reliable semantics of transport layer multicast and define three classes of reliable semantics: best effort, all-reliable and statistical-reliable. Specifically, we investigate the problem posed by the reliable service of the XTP multicast (we refer to such a service as statistical reliable) which is primarily related to error recovery issues. XTP error control scheme uses a bucket mechanism kept at the sender side. We discovered that the XTP multicast response to errors is inefficient. This drawback is caused by the multiple retransmission of the same missing data packets. We show that the total number of redundant retransmissions may be proportional to the number of buckets. To overcome this problem, we introduce soem other constraints in the sender side to prevent any unnecessary retransmission.
Keywords: XTP; transport layer; multicast; error control

Nachum Shacham, "Multicast routing of hierarchical data," in Conference Record of the International Conference on Communications (ICC), (Chicago, Illinois), pp. 1217-1221, IEEE, June 1992.

Abstract: Multicast of real-time data in a heterogeneous network environment, in which links and recipients differ in their bandwidth, is considered. Traditional schemes are designed to deliver the source's complete data to all recipients, thereby restricting the source's handling of heterogeneity to either overcompressing to the bandwidth accessible to the least capable user, or excluding those destinations who cannot receive the full signal. We present an alternative approach, in which the source encodes its signal hierarchically, and the network delivers subsets of the signal layers in accordance with individual destinations' bandwidth constraints. This approach allows each user to individually trade off reception bandwidth for signal quality independent of other users' selections. In support of such service, efficient routes that carry the desired bandwidth to all destinations must be computed. Routing algorithms that compute such paths are presented, and their relative merits are compared.
Keywords: multicast; routing; hierarchical coding; multimedia

Nachum Shacham, "Multipoint communication by hierarchically encoded data," in Proceedings of the Conference on Computer Communications (IEEE Infocom), vol. 3, (Florence, Italy), pp. 2107-2114 (9A.4), IEEE, May 1992.

Abstract: Many types of signals, such as video and speech, allow a recipient to extract essential information even when receiving only portions of the signal; and users, under terminal or network-access constraints, would rather receive partial information than no information at all. We present a new multipoint communication paradigm, in which each destination receives a subset of the source's signal that corresponds to that destination's terminal and access bandwidth constraints. The approach to realizing this paradigm is based on integration of layered coding of the source's signal, routing based on bandwidth demand, optimization of signal parameters, and layered error control. In this paper, we overview several hierarchical signal coding techniques; and present methods for finding maximum bandwidth available to destinations, establishing maximum-bandwidth routes; and optimally assign bandwidth to the signal layers to maximize overall reception quality. We also present error control procedures whereby the network, source, and destinations cooperate to maintain layered-based data integrity, using erasure-recovery coding and prioritized packet deletion.
Keywords: error control; layered coding; packet voice; packet video; multicast

K. J. Schrodi, B. Pfeiffer, J. M. Delmas, and M. De~Somer, "Multicast handling in a self-routing switch architecture," in International Switching Symposium, vol. 2, (Yokohama), p. A7.1, Oct. 1992.

Abstract: This paper discusses the requirements for multicast support in future ATM based switching networks and it gives an overview of some key architectural options for implementing the related capabilities in multipath self-routing switching network architectures. An actual solution is described, which has been implemented in the Multipath Self-Routing (MPSR) switch, a highly flexible, performant and fault tolerant switching concept.
Keywords: Multicast; self routing; switching; switching network; architecture; ATM; multipath

H. Tode, Y. Sakai, M. Yamamoto, H. Okada, and Y. Tezuka, "Multicast Routing Algorithm for Nodal Load Balancing," in Proceedings of the Conference on Computer Communications (IEEE Infocom), (Florence), pp. 2086-2095, 1992.

Abstract: In this paper we propose two multicast routing algorithms which distribute copy operation of packets over all nodes along the multicast path: a link added type algorithm and a loop construc-ted type algorithm. Both algorithms, at first, derive an approximate solution for minimum cost path, and then improve thesolution to prevent concentration of packet copy operation at one switching node at a little sacrifice of total cost along the
Keywords: Multicast; routing algorithm; performance evaluation; simulation

Dinesh C. Verma, "Routing reserved bandwidth multi-point connections," in SIGCOMM Symposium on Communications Architectures and Protocols (Deepinder P. Sidhu, ed.), (San Francisco, California), pp. 96-105, ACM, Sept. 1993. also in \em Computer Communication Review 23 (4), Oct. 1992.

Abstract: Some important classes of multi-point bandwidth-intensive applications like video-conferencing with mixing and the distributed classroom can be characterized as consisting of a broadcast from a source node to several destination nodes, and point-to-point flows from the destination nodes to the source node. Determining a tree in an arbitrary mesh network which satisfies the bandwidth constraints and minimizes the cost of reserved bandwidth is an NP-hard problem. In this paper, we look at some heuristics that can be used to solve the problem of routing these multi-point connections. The heuristics are based on finding the capacity-constrained minimum cost tree which minimizes the cost of bandwidth reserved for point-to-point communication from destinations to the source, and weights are assigned to minimize the number of extra nodes in the tree which increase the cost of bandwidth reserved from the source to the destination. A theoretical bound on the performance of some of the heuristics, as well as simulation results comparing their performance to that of the optimum solution are presented. The results are encouraging, the heuristics find a tree with a cost within 2% of the optimum on average, and with a cost within 10% of the optimum in those cases when the heuristic fails to find the optimum tree.
Keywords: routing; multicast; spanning tree; optimization

R. Venkatesan and H. T. Mouftah, "Performance Analysis of Multipath Banyan Networks," in Conference Record of the International Conference on Communications (ICC), pp. 912-916, 332.6.1-332.6.5, June 1992.

Abstract: The performance of the replicated and dilated banyan networks, which are being considered for use in broadband packet switch architectures, is analyzed. A strong arguement is made for comparing the performances of the various multipath networks using the assumption that multiple packets can be accepted at any destination. Multipath banyan networks with one internal buffer in each switching element are also studied. [...] An ATM switch should posses a very high throughput (several gigabits per second), a low switching delay (about half a millisecond), a low packet loss probability, expandability, testability and fault tolerance, low cost and flexibility to achieve broadcasting as well as multicasting. Accordingly, the network structure used in the lowest level of an ATM switch should posses a very high unbuffered probability of acceptance; besides, a simple distributed routing scheme is desirable. Multipath feature is a must if broadcasting and multicasting are to be made feasible. Scores of different multistage interconnection networks (MINs) have been reported to date, varying in their complexity, throughput, flexibility, robustness, fault tolerance and other properties [1,2,3]. Most of the MINs studied for broadband communication are delta networks, and quite frequently they are based on 2x2 SEs [6,7,8,9]. In the analysis of MINs we usually assume that only one packet can be accepted at any destination. While this assumption is ideal for comparing the performance of unique path networks, we do not obtain a good indication of multipath networks. In this paper, we analyze the performances of multipath banyan networks assuming that any number of packets may be accepted at any output stage in a given cycle. In addition, performances of multipath banyan networks with one internal buffer in each switching element are studied.; references=; [1] T.Y. Feng, "A survey of interconnection networks," IEEE Computer, vol. 14, no. 12, pp. 12-27, Dec. 1981. [2] H.J. Siegel, Interconnection networks for large scale parallel processing; theory and case studies, Second Edition, McGraw Hill, 1990. [3] S. Sundaram and R. Venkatesan, "Reconfigurable multistage interconnection networks; a state of the art review," MUN EAST Report no. 89-001, Aug. 1989. [6] D.M. Dias and J.R. Jump, "Analysis and simulation of buffered delta networks," IEEE Transactions on Computers, vol. C-30, no. 4, pp. 273-282, Apr. 1981. [7] Y.C. Jenq, "Performance analysis of a packet switch based on a single-buffered banyan network," IEEE Journal of Selected Areas in Communications, vol. SAC-3, no. 6, pp. 1014-1021, Dec. 1983. [8] H. Yoon, K.Y. Lee and M.T. Liu, "Performance analysis of multibuffered packet switching networks in multiprocessor systems," IEEE Transaction on Computers, vol. C-39, no. 3, pp. 319-327, Mar. 1990. [9] H.F. Badran, and H.T. Mouftah, "Performance of output-buffered broadband switch architectures," Proceedings of Canadian Conference on Electrical and Computer Engineering, Ottawa, pp. 39.2.1-39.2.4, Sep. 1990.
Keywords: ATM banyan network, multipath multistage interconnection network (MIN) crossbar network switch architecture banyan network, 4-dilated performance analysis

A. G. Waters, "Multicast provision for high speed networks," in 4th IFIP Conference on High Performance Networking, (Liège, Belgium), pp. G1-1 - G1-16, IFIP, Dec. 1992.

Abstract: With the advent of high speed packet networks capable of supporting integrated services, there is increasing scope for multi-party applications such as multimedia conferencing, file distribution, video distribution, real-time data collection and computer supported cooperative work. Such applications need to be properly supported by the network in order to encourage their development. This paper presents a layered framework offering the necessary support mechanism at appropriate levels to make effective and efficient use of networking resources. Two aspects of multicast provision are then dealt with in more detail: first, multicast routing for networks of arbitrary topology and secondly, a flexible set of group-management procedures.
Keywords: multicast; routing; group management; video conferencing; conference control; Steiner tree

L. A. Wang and K. C. Lee, "A WDM Based Virtual Bus for Universal Communication and Computing Systems," in Conference Record of the International Conference on Communications (ICC), pp. 888-894, 332.2.1-332.2.7, June 1992.

Abstract: We propose a scalable interconnection network architecture called virtual bus to solve the diverse and non-uniform traffic encountered in a universal system. We exploit the feasibility of using wavelength division multiplexing (WDM) technologies to reduce the wiring complexity and the power consumption of the virtual bus. The scalability and broadcasting capability of the virtual bus is achieved by hierarchical clustering of WDM buses and by interconnecting the WDM clusters via highly parallel VLSI based switches. We show that a high performance WDM virtual bus design sustaining giga packets and tera bits per second non-uniform traffic can be implemented efficiently.
Keywords: ATM switch design switch architecture WDM optical bus multicast interconnection network

Liming Wei, FongChing Liaw, Deborah Estrin, Allyn Romanow, and Tom Lyon, "Analysis of a resequencer model for multicast over ATM networks," in Third International Workshop on network and operating system support for digital audio and video, (San Diego, California), pp. 197-208, IEEE Communications Society, Nov. 1992.

Abstract: Multicast delivery saves bandwidth and offers logical addressing capabilities to the applications. The receivers of a multicast group need to differentiate cells sent by different sources. This demultiplexing requirement can be satisfied in an ATM environment using multiple dedicated point-to-multipoint virtual channel connections (VCs), but with certain shortcomings. This paper discusses an alternative resequencing model to solve this problem. It scales well in large networks. Three resequencing methods are developed and simulation results reported. The strategy is useful for applications spanning large regions where it is desirable to mix streams of cells from different bursty sources onto the same virtual channel.
Keywords: multimedia; multicast; ATM; resequencing
Annotation: A designated source in the group is elected as a resequencer. All other sources send their multicast ATM cells to the designated resequencer. Members in each region elect a source to act as a resequencer.

P. C. Wong and M. S. Yeung, "Pipeline Banyan - A Parallel Fast Packet Switch Architecture," in Conference Record of the International Conference on Communications (ICC), pp. 882-887, 332.1.1-332.1.6, June 1992.

Abstract: In this paper, we propose a new fast packet switch architecture. This switch has a control plane and a number of parallel data planes which are of the same banyan topology. Packet headers are routed by the control plane to set the corresponding routing paths in the data planes. Since the data planes to not need to do routing decisions, their hardware complexity can be reduced. The Pipeline Banyan has output queueing capability and can approach 100% maximum throughput. It can deliver packets in a sequential order. The internal switching speed needs only be a fraction of the input port speed. The basic structure is a regular NxN Banyan which is suitable for VLSI implementation.
Keywords: ATM switch design switch architecture multicast broadcast banyan, buffered switch, Batcher-Banyan banyan, pipline simulation

W. D. Zhong, S. Shimamoto, Y. Onozato, and J. Kaniyil, "A recursive copy network for a large multicast ATM switch," in International Switching Symposium, vol. 2, (Yokohama), p. A7.2, Oct. 1992.

Abstract: This paper presents a new architecture of a copy network which employs the principle of recursion for the replication of cells. The proposed architecture achieves high utilization of the links and buffers of the copy network. Two different modular structures of large multicast ATM switches are configured.
Keywords: ATM; multicast; network; architecture; simulation