Search

New real-time applications frequently involve timing constraints related to accurate services from communication protocols. Concretely, real-time communication protocols utilize timers to implement these constraints between system event occurrences. In this context, the study of congestion control for Internet reliable multicast is at present an active research field related to real-time protocols. In this paper, the authors present an innovative real-time transport protocol named Scalable Reliable Multicast Stair Hybrid (SRMSH) as new hybrid multiple layer mechanism for multicast congestion control providing detection and recovery loss. This work is focused on formal specification of SRMSH approach using Communicating Real-Time State Machines as a formal method. Besides, SRMSH validation is presented within a formal proof framework in order to check the functional safety and liveness properties. As a result, authors outline a dynamical system framework in order to model behavior of their presented solution.

Transport level multicast protocols providing reliability and scalability properties are certainly essential building blocks for several distributed group applications. We consider the effect of reliable multicast transport mechanisms on traffic characteristics and hence network performance. Although self-similarity property of unicast traffic, in particular TCP, has been analyzed extensively, multicast traffic has not been incorporated from this perspective. In this study, we focus on traffic characterization of transport level reliable multicasting. In particular, we concentrate on two scalable and reliable multicast protocols as case studies, namely Bimodal Multicast and Scalable Reliable Multicast (SRM), and analyze the traffic generated by them. Our study consists of a complete simulation analysis supported by theoretical work, which shows that self-similarity is protocol dependent. We demonstrate that the Markovian character of Bimodal Multicast's epidemic loss recovery distinguishes an inherently superior protocol. It discretely feeds well-behaved traffic and copes with the existing self-similarity. On the other hand, the feedback controlled loss recovery mechanism of SRM triggers self-similarity. Drawing upon both theoretical and simulation analysis, our results substantiate that transport level can induce long-range dependence even in the absence of application/user level causes.

In this paper, we propose the design of a scalable reliable mobile multicast scheme-SRMoM. SRMoM uses well-known Scalable Reliable Multicast (SRM) in the wired networks and a NAK-based ARQ with adaptive Forward Error Correction (AFEC) in the wireless networks. In AFEC, the probability of needing retransmission of original multicast packets after FEC recovery is selectable. This selective property enables the control of channel utilization in the wireless segment for different numbers of Mobile Hosts (MHs). Using this property, the channel utilization of SRMoM is made to be virtually independent of the number of MHs, thus making it extremely scalable. The performance of SRMoM is analyzed with three adaptive FEC algorithms based on three wireless loss models, namely a Gilbert-Elliott channel, a simplified Cilbert-Elliott channel, and a binary symmetric channel, analytically as well as through simulation. Furthermore, the performance of SRMoM is compared with SRM and MRMoM (NAK-based protocol without FEC) through simulation. Using the average number of transmissions per original multicast packet, and wireless link utilization as metrics, we demonstrate that the performance of SRMoM is indeed virtually independent of the number of MHs, and that it results in the lowest number of packet transmissions and lowest channel utilization of reliable mobile multicast protocols thai have been proposed to date.

Traditional reliable multicast protocols depend on assumptions about flow control and reliability mechanisms, and they suffer flom a kind of interference between these mechanisms. This in turn affects the overall performance, throughput and scalability of group applications utilizing these protocols. However, there exists a substantial class of distributed applications for which the throughput stability and scalability guarantees are indispensable. Bimodal Multicast (Pbcast) is a new option in scalable reliable multicast protocols that uses an inverted protocol stack approach, in which probabilistic mechanisms are used at low layers, and reliability properties introduced closer to the application. The main contributions of this study are development of simulation models for performance evaluation of Bimodal Multicast, demonstration of how the inverted protocol stack approach works well on several network settings, and its comparison with best-effort reliable multicast mechanisms. Analysis results reveal that Bimodal Multicast, together with optimizations for improving its latency and reliability characteristics, scales well, exhibits stable throughput and in contrast to the other scalable reliable multicast mechanisms it gives predictable reliability even under highly perturbed conditions.

We examine an approach for providing reliable, scalable multicast communication, involving the use of multiple multicast channels for reducing receiver processing costs and reducing network bandwidth consumption in a multicast session. In this approach a single multicast channel is used for the original transmission of packets. Retransmissions of packets are done on separate multicast channels, which receivers dynamically join and leave. We first show that protocols using an infinite number of multicast channels incur much less processing overhead at the receivers compared to protocols that use only a single multicast channel. This is due to the fact that receivers do not receive retransmissions of packets they have already received correctly. Next, we derive the number of unwanted redundant packets at a receiver due to using only a finite number of multicast channels, for a specific negative acknowledgment (NAK)-based protocol. We then explore the minimum number of multicast channels required to keep the cost of processing unwanted packets to a sufficiently low value. For an application consisting of a single sender transmitting reliably to many receivers we find that only a small number of multicast channels are required for a wide range of system parameters. In the case of an application where all participants simultaneously act as both senders and receivers a moderate number of multicast channels is needed. Finally, we present two mechanisms for implementing multiple multicast channels, one using multiple IP multicast groups and the other using additional router support for selective packet forwarding. We discuss the impact of both mechanisms on performance in terms of end-host and network resources.

There are many methods for making a multicast protocol “reliable.” At one end of the spectrum, a reliable multicast protocol might offer tomicity guarantees, such as all-or-nothing delivery, delivery ordering, and perhaps additional properties such as virtually synchronous addressing. At the other are protocols that use local repair to overcome transient packet loss in the network, offering “best effort” reliability. Yet none of this prior work has treated stability of multicast delivery as a basic reliability property, such as might be needed in an internet radio, television, or conferencing application. This article looks at reliability with a new goal: development of a multicast protocol which is reliable in a sense that can be rigorously quantified and includes throughput stability guarantees. We characterize this new protocol as a “bimodal multicast” in reference to its reliability model, which corresponds to a family of bimodal probability distributions. Here, we introduce the protocol, provide a theoretical analysis of its behavior, review experimental results, and discuss some candidate applications. These confirm that bimodal multicast is reliable, scalable, and that the protocol provides remarkably stable delivery throughput.

Reliable multicast communication is important in large-scale distributed applications. For example, reliable multicast is used to transmit terrain and environmental updates in distributed simulations. To date, proposed protocols have not supported these applications' requirements, which include wide-area data distribution, low-latency packet loss detection and recovery, and minimal data and management over-head within fine-grained multicast groups, each containing a single data source.In this paper, we introduce the notion of Log-Based Receiver-reliable Multicast (LBRM) communication, and we describe and evaluate a collection of log-based receiver reliable multicast optimizations that provide an efficient, scalable protocol for high-performance simulation applications. We argue that these techniques provide value to a broader range of applications and that the receiver-reliable model is an appropriate one for communication in general.

Many networking applications such as video-on-demand services, video conferencing, and distributed database replications require multicast communications. Providing scalable and reliable multicast are basic requirements to develop a better routing protocols which is used for disaster management, emergency relief, mobile conferencing among many other applications. As mobile host changes its access point over a time interval, multicast routes have to be updated frequently. This poses several challenges to provide an efficient multicast routing. In this paper, we propose a Scalable Reliable Multicast Ad hoc On Demand Distance Vector (SRMAODV) protocol based on Ad hoc On Demand Distance Vector (AODV) and Multicast AODV (MAODV) protocol to support scalability and reliability for online routing of delivery-guaranteed multicasts. Exhaustive simulation experiments reveal that Packet Delivery Ratio (PDR) which is an important measure of reliability for SRMAODV is much better together with improved end-to-end delay.

Content Delivery Networks (CDNs) provide an efficient support for serving http and streaming media content while minimizing the network impact of content delivery as well as overcoming the server overload problem. Not too much research effort has been placed in the analysis of the access network and the impact and benefits of multicast. This paper proposes an innovative real-time protocol offering an hybrid congestion control algorithm called Scalable Reliable Multicast Stair Hybrid (SRMSH) which combines the benefits of TCP AIMD with Rate-based (STAIR) [2] for layered multicast with Scalable Reliable Multicast (SRM) [6] in order to offer loss recovery methods if receivers detect loss events. The mechanism has been simulated and implemented to proof its benefits.

With the increasing deployment of multimedia real-time Internet applications, evaluating transport protocol metrics of quality of service (QoS) has gained rapidly increasing importance. In this paper, a novelty protocol named scalable reliable multicast stair hybrid (SRMSH) is presented as new hybrid multiple layer mechanism for multicast congestion control providing detection and recovery loss. Then SRMSH is simulated with video streaming traffic source to measure fundamental components to real-time multimedia applications: throughput, latency and jitter. Work is focused on performance analysis and results obtained from NS-2 traces clearly conclude that SRMSH exhibits interesting insights using these metrics of real-time multimedia applications

Due to the distribution nature and multi-customer interaction characteristic of the CSCW system, multicast becomes an ideal transmission instrument. Thus we have put into effect a multicast protocol named TORM (totally ordered reliable multicast). In order to evaluate its performance, we conduct the analysis and comparison between TORM and another multicast protocol, SRM (scalable reliable multicast), in respect of their error recovery mechanism. With queuing theory, we establish a queue model, based on which the expression of error recovery latency are worked out. The analytical result demonstrates TORM with the higher error recovery efficiency is often preferable in the CSCW application. Moreover, we simulate the two protocols with NS2, a network simulation tool, and get the same conclusion.

The study of congestion control for Internet reliable multicast is at present an active research field. In this paper we present scalable reliable multicast stair hybrid (SRMSH) as a multiple layer mechanism for multicast congestion control providing detection and recovery loss. We simulated SRMSH with different traffic sources to measure heterogeneity, scalability and fairness. Our work focus on bandwidth analysis and results clearly conclude that SRMSH exhibits reasonable throughput using these metrics and a good performance competing with typical traffic flow from Internet like TCP.

Group communications (multicast) are foreseen to be one of the most critical yet challenging technologies to meet the exponentially growing demands for data distribution in a large variety of applications of the Internet (such as grid computing, web applications and distributed simulations). When reliability is required, there is no straightforward solutions and meeting the objectives of reliable multicast is not an easy task. Active networks open a new perspective in providing more efficient solutions for the problem of reliability. In this context, routers are able to perform customized computations on the packets flowing through them. In this report, we propose a receiver-based (replier) local recovery multicast protocol with dynamic repliers elected on a per-packet basis. Designed to provide an efficient reliable multicast service without any cache facilities inside the network, our approach, uses low-overhead active services in routers. The current report addresses the design, evaluation and the implementation of an efficient and scalable reliable multicast protocol noted DyRAM standin for Dynamic Replier Active reliable Multicast.

Scalable reliable multicast protocols have been the focus of recent research, tackling the problem of efficient reliable data delivery to an arbitrarily large number of receivers. Yet, the common applications of multicast, such as multi-point file delivery, or video streaming from a media server, typically only involve a moderate number of receivers, such as a thousand or fewer. Moreover, because of the limited deployment of these specialized multicast protocols, it is common, when feasible, for applications to use multiple TCP connections instead, one for each receiver, to implement multi-point delivery, causing a significant demand on the transmission server and the downstream links. We describe a multicast extension to TCP, called single-source multicast optimization (SMO), that optimizes this case of multipoint delivery, providing the benefits of multicast together with the familiar features and API of TCP. Our results from experiments based on a Linux implementation and performed on a testbed show that TCP-SMO requires just a modest extension to the TCP implementation and provides scalable performance of multicast up to over a thousand receivers, thereby satisfying the common case requirements. In addition, used with TCP-RTM (real-time mode), TCP-SMO also supports real-time multimedia multicast applications well.

As the scale of a network becomes larger in terms of both the number of users and geographic span, reliable multicast protocols suffer more severely from acknowledgement implosion and exposure to retransmissions. In this paper, we propose a scalable, efficient recovery scheme for a tree-based reliable multicast transport protocol. The scheme first constructs a logical tree of receivers as close to a multicast routing tree as possible by using error bitmap information. This ensures that the receivers residing at a higher level in a multicast routing tree than ones requesting retransmissions be appointed as parents in a corresponding logical tree. In our scheme, the logical tree is adaptively reconfigured as the session membership or the multicast route changes. Our scheme also forms local groups with separate multicast addresses in the tree for efficient recovery. We compare our scheme with a static tree-based reliable multicast protocol. Results of the simulation show that our adaptive tree-based recovery scheme outperforms the compared protocol in terms of the implosion and exposure as the session size increases.

The development of high-speed networks and the expansion of the Internet have increased both geographical extent and participant population of applications such as videoconferencing, multimedia dissemination, electronic stock exchange, and distributed cooperative work. The key property of this type of applications is the need to distribute data among multiple participants together with application specific quality of service needs which fact makes multicast protocols an essential underlying communication structure. In this paper, we analyze traffic characteristics of two scalable multicast protocols, namely Bimodal Multicast (Pbcast) and Scalable Reliable Multicast (SRM), each having different approaches for loss recovery and providing reliability. Particularly, our simulation studies demonstrate that epidemic approach of Bimodal Multicast generates a more desirable traffic than SRM with lower overhead traffic and transport delays. SRM delays show long-range dependence and self-similarity whereas Bimodal Multicast delays are shortrange dependent. Self-similarity and long-range dependence are ubiquitous in wide area networks, which lead to adverse consequences in network performance. We elaborate on the protocol mechanisms as the underlying factor in our empirical results. The intrinsic relation of these mechanisms to traffic characteristics is explored.

We analyze the maximum throughput that the known classes of reliable multicast protocols can attain. A new taxonomy of reliable multicast protocols is introduced based on the premise that the mechanisms used to release data at the source after correct delivery should be decoupled from the mechanisms used to pace the transmission of data and to effect error recovery. Receiver-initiated protocols, which are based entirely on negative acknowledgments (NAKs) sent from the receivers to the sender, have been proposed to avoid the implosion of acknowledgments (ACKs) to the source. However, these protocols are shown to require infinite buffers in order to prevent deadlocks. Two other solutions to the ACK-implosion problem are tree-based protocols and ring-based protocols. The first organize the receivers in a tree and send ACKs along the tree; the latter send ACKs to the sender along a ring of receivers. These two classes of protocols are shown to operate correctly with finite buffers. Following our taxonomy, the maximum attainable throughput by the known classes of reliable multicast protocols is analyzed. It is shown that tree-based protocol constitute the most scalable class of all reliable multicast protocols proposed to date.

The success of multicast applications such as Internet teleconferencing illustrates the tremendous potential of applications built upon wide-area multicast communication services. A critical issue for such multicast applications and the higher layer protocols required to support them is the manner in which packet losses occur within the multicast network. We present and analyze packet loss data collected on multicast-capable hosts at 17 geographically distinct locations in Europe and the US and connected via the MBone. We experimentally and quantitatively examine the spatial and temporal correlation in packet loss among participants in a multicast session. Our results show that there is some spatial correlation in loss among the multicast sites. However, the shared loss in the backbone of the MBone is, for the most part, low. We find a fairly significant amount of of burst loss (consecutive losses) at most sites. In every dataset, at least one receiver experienced a long loss burst greater than 8 seconds (100 consecutive packets). A predominance of solitary loss was observed in all cases, but periodic losses of length approximately 0.6 seconds and at 30 second intervals were seen by some receivers.

Retransmission based error control for large scale multicast applications is difficult because of implosion and exposure. Existing schemes (SRM, RMTP, TMTP LBRRM) have good solutions to implosion, but only approximate solutions to exposure. We present a scheme that achieves finer grain fault recovery by exploiting new forwarding services that allow us to create a dynamic hierarchy of receivers. We extend the IP multicast service model so that routers provide a more refined form of multicasting (which may be useful to other applications), that enables local recovery. The new services are simple to implement and do not require routers to examine or store application packets; hence, they do not violate layering. Besides providing better implosion control and less exposure than other schemes, our scheme integrates well with the current IP model, has small recovery latencies (it requires no back-off delays), and completely isolates group members from topology. Our scheme can be used with a variety of multicast routing protocols, including DVMRP and PIM. We have implemented our scheme in NetBSD Unix, using about 250 lines of new C-code. The implementation requires two new IP options, 4 additional bytes in each routing entry and a slight modification to IGMP reports. The forwarding overhead incurred by the new services is actually lower than forwarding normal multicast traffic.

We have developed a scalable reliable multicast architecture for delivering one-to-many telepresentations. In contrast to audio and video, which are often transmitted unreliably other media, such as slides, images and animations require reliability. Our approach transmits the data in two layers. One layer is for session-persistent data, with reliability achieved by FEC alone, using the Fcast protocol. The other layer is for dynamic data, with reliability achieved using the ECSRM protocol, which combines FEC with NACK suppression. Our approach is scalable to large heterogeneous receiver sets, and supports late-joining receivers. We have implemented our approach in a multicast version of PowerPoint, a graphical slide presentation tool.

Local recovery approaches for reliable multicast have the potential to provide significant performance gains in terms of reduced bandwidth and delay, and higher system throughput. In this paper we examine two local recovery approaches-one server-based, and the other receiver-based, and compare their performance. The server-based approach makes use of specially designated hosts, called repair servers, co-located with routers inside the network. In the receiver-based approach, only the end hosts (sender and receivers) are involved in error recovery. Using analytical models, we first show that the two local recovery approaches yield significantly higher protocol throughput and lower bandwidth usage than an approach that does not use local recovery. Next, we demonstrate that server-based local recovery yields higher protocol throughput and lower bandwidth usage than receiver-based local recovery when the repair servers have processing power slightly higher than that of a receiver and several hundred kilobytes of buffer per multicast session.

Reliable multicast protocols on top of the MBone are presently subject to intensive research. In the past, numerous protocols have been developed and their respective performance been analysed. Little progress has been made, though, to compare different approaches. We use the network simulator ns-2 to evaluate the performance of three protocols, namely scalable reliable multicast (SRM), multicast file transfer protocol (MFTP) and an enhanced version of the latter, called multicast file transfer protocol with erasure correction (MFTP/EC). We also compare the results to each other and test the suitability for multicast file distribution.

End-to-end protocols measure network characteristics and react based on their estimates of network performance. Network dynamics can alter the topology significantly, and thereby affect protocol operation. Topology changes may result in routing pathologies (such as route loops, packet interleaving), changes to the end-to-end path characteristics, network partition etc., that then impact the performance of end-to-end protocols. This paper presents methodologies to evaluate an end-to-end protocol in the presence of network dynamics using a simulator. We evaluate a reliable multicast transport protocol over dynamic topologies and study its adaptivity to topology change. We present a systematic evaluation of the adaptive timer mechanisms in scalable reliable multicast (SRM). The timer mechanisms are evaluated under simple topology changes, as well as under network partition conditions. The paper concludes by posing a number of open research questions about the behaviour of different reliable multicast mechanisms when operating over dynamic topologies.

There are many scenarios in which the same data must be delivered over a packet switched network to a large set of receivers. The Internet enables efficient multipoint transmissions through IP multicast by allowing data transmission to all receivers with a single send. Most approaches to scalable reliable multicast utilize receiver-oriented retransmissions. Defining an API for receiver-oriented reliable multicast is difficult because it is not clear how to manage the sender's cache and to schedule repairs. We outline an approach to defining an API based on logical cache persistence that addresses these problems. We also we explore the issues involved in defining an API for reliable multicast protocol on the Internet that can scale to millions of receivers.

Error recovery by retransmission for large scale multicast faces two intrinsic scalability challenges: acknowledgment implosion and exposure to retransmission. This paper proposes a new scheme to reduce the cost of error recovery by using the idea of cumulative negative acknowledgment (CNAK). In the scheme, we assume a window control of size W. A receiver does not send back a NAK immediately upon the discovery of a packet loss. Instead, the receiver accumulates the losses and returns a CNAK to the source at about half of the window interval. Retransmission is carried out on multiple multicast channels which are mapped to window size W. The A /sub i/ multicast channel is mapped to the sequence number i where i = 1,2,...,W. Our scheme has these advantages: one CNAK replaces W/2 number of conventional NAK, thus, greatly reducing the NAK overhead. Moreover, the use of multiple multicast channels for packet retransmission completely eliminates the exposure hazard. We apply the CNAK scheme to the DR local recovery model and show that the proposed C4 protocol is the most efficient protocol among all. The protocol is highly scalable and free of unwanted repair packets.

The round trip network delay times between pairs of nodes in a multicast session is a key parameter; it is used in suppressing the implosion of request and repair packets, in detecting congestion, etc. This paper presents a protocol to estimate the delays between every pair of nodes in a multicast group. While existing protocols use O(n) multicast messages with O(n) size each (total of O(n/sup 2/) bits), our protocol requires O(n) multicasts with only O(1) size each (total of O(n) bits); here, n is the session size. It does not require clocks to be synchronized, or any knowledge of network topology and the session size. We incorporate our protocol into the SRM (scalable reliable multicast) protocol (Rhee et al., 1999) and show by our simulations that, (i) our delay estimation protocol is reasonably accurate, (ii) it is just as effective as existing delay estimation protocols in suppressing repair and request traffic in SRM, and (iii) it has significantly lower overhead than the existing delay estimation protocols. Furthermore, our protocol can be extended to run within a hierarchically structured multicast protocol such as RMTP (reliable multicast transport protocol) (Paul et al., 1996). Here the goal is to estimate round trip delays from a sender to each receiver. Our protocol ensures that only the immediate children of the sender send feedback messages directly to the sender, thereby avoiding feedback implosion. Our delay estimation protocols therefore significantly enhance the scalability of unstructured multicast protocols such as SRM, as well as tree-structured multicast protocols such as RMTP.

Many scalable reliable multicast protocols use the local repair scheme where certain receivers retransmit packets by other receivers. Such schemes need a mechanism, called message stability, to ensure reliable delivery to all members of a multicast group and to delete those packets received by all members from the buffers of the local repairers. We propose a new protocol for message stability based on random gossiping. The protocol offers scalabilty and fault-tolerance by limiting each of its message transmissions only to a constant number of randomly chosen group members, hence eliminating message implosion and single point failure through the diffusion of responsibility. Both statistical analysis and simulation study indicate that our gossip-style message stability protocol can be highly effective for large-scale reliable multicast.

In this paper, we present a formal model of the reliable multicast service that ensures eventual packet delivery with, possibly, some timeliness guarantees. This model dictates precisely what it means to be a member of the reliable multicast group and which packets are guaranteed delivery to which members of the group. Moreover, it is reasonable, implementable, and broad; that is, it captures the intended behavior of numerous reliable multicast protocols. We also present a formal model of the Scalable Reliable Multicast (SRM) protocol [1]. We show that our model of SRM is safe, in the sense that it is a faithful implementation of our model of the reliable multicast service; that is, it may only deliver appropriate packets to each member of the reliable multicast group. We also show that, under certain constraints, the implementation is live, in the sense that it guarantees the timely delivery of the appropriate packets to the appropriate members of the reliable multicast group.

Traditional reliable multicast protocols depend on assumptions about flow control and reliability mechanisms, and they suffer from a kind of interference between these mechanisms. This in turn affects the overall performance, throughput and scalability of group applications utilizing these protocols. However, there exists a substantial class of distributed applications for which the throughput stability guarantee is indispensable. Pbcast protocol is a new option in scalable reliable multicast protocols that offers throughput stability, scalability and a bimodal delivery guarantee as the key features. In this paper, we focus on the throughput stability of reliable multicast protocols. We describe an experimental model developed for Pbcast and virtually synchronous protocols on a real system. We then give the analysis results of our study.

We have developed a parallel software-only processing system for creating real-time video effects such as titling and compositing (e.g., picture-in-picture) using compressed Internet video sources. The system organizes processors into a hierarchy of levels. Processes at each level of the hierarchy can exploit different types of parallelism and coordinate the actions of lower levels. To control the effect, control messages must be distributed to processors in the hierarchy while preserving the independence of each level. This requires a control mechanism that supports efficient delivery of messages to groups of processors, tunable reliability semantics, and recoverable state information. We describe a mechanism that meets these requirements that uses IP-Multicast, the Scalable Reliable Multicast protocol, and the Scalable Naming and Announcement Protocol. We also describe an optimization that provides a flexible framework for linking the control of different aspects of one or more related video effects.

In this paper, we propose a transport layer protocol for oneto-many multicast that is designed to have good scalability properties in large receiver groups and provides reliable transmission. Reliability is achieved using forward error correction (FEC) in combination with selective repeat NAK-based ARQ. Using FEC can significantly reduce the necessity for retransmission requests or make them totally unnecessary, which is important in wireless and satellite communicaton environments, as well as for delay-sensitive multimedia applications. The protocol uses one multicast group address for the original transmission and a second group address for the handling of retransmissions, which helps in significantly reducing the network load on branches with low loss and facilitates usage in the context of satellite communications. We hope that our work will be useful and encouraging for the development of group communication applications.

To deliver multicast messages reliably in a group, each member maintains copies of all messages it sends and receives in a buffer for potential local retransmission. The storage of these messages is costly and buffers may grow out of bound. Garbage collection is needed to address this issue. Garbage collection occurs once a process learns that a message in its buffer has been received by every process in the group. The message is declared stable and is released from the process''s buffer. This paper proposes a gossip-style garbage collection scheme called GSGC for scalable reliable multicast protocols. This scheme achieves fault-tolerance and scalability without relying on the underlying multicast protocols. It collects and disseminates information in the multicast group by making each group member periodically gossip information to a random subset of the group. Extending the global gossip protocol further, this paper also investigates a local gossip scheme that achieves improved scalability and significantly better performance. Simulations conducted in a WAN environment are used to evaluate the performance of both schemes.

We examine an approach for providing reliable, scalable multicast communication, using multiple multicast groups for reducing receiver processing costs in a multicast session. In this approach a single multicast group is used for the original transmission of packets. Retransmissions of packets are done to separate multicast groups, which receivers dynamically join or leave. We first show that by using an infinite number of multicast groups, processing overhead at the receivers are substantially reduced. Next, we show that, for a specific negative acknowledgment (NAK)-based protocol, most of this reduction can be obtained by using only a small number of multicast groups for a wide range of system parameters. Finally, we present a local filtering scheme for minimizing join/leave signaling when multiple multicast groups are used.

JetFile is a file system designed with multicast as its distribution mechanism. The goal is to support a large number of clients in an environment such as the Internet where hosts are attached to both high and low speed networks, sometimes over long distances. JetFile is designed for reduced reliance on servers by allowing client-to-client updates using scalable reliable multicast. Clients on high speed networks prefetch large numbers of files. On low speed networks such as wireless, special caching policies are used to decrease file access latency. The prototype implementation of JetFile is on the JetStream gigabit local area network which provides hardware support for many multicast addresses. The multicast Internet backbone (Mbone) is the wide area testbed for JetFile.

State of the art, real-time, rate-adaptive, multimedia applications adjust their transmission rate to match the available network capacity. Unfortunately, this source-based rate-adaptation performs poorly in a heterogeneous multicast environment because there is no single target rate --- the conflicting bandwidth requirements of all receivers cannot be simultaneously satisfied with one transmission rate. If the burden of rate-adaption is moved from the source to the receivers, heterogeneity is accommodated. One approach to receiver-driven adaptation is to combine a layered source coding algorithm with a layered transmission system. By selectively forwarding subsets of layers at constrained network links, each user receives the best quality signal that the network can deliver. We and others have proposed that selective-forwarding be carried out using multiple IP-Multicast groups where each receiver specifies its level of subscription by joining a subset of the groups. In this paper, we extend the multiple group framework with a rate-adaptation protocol called Receiver-driven Layered Multicast, or RLM. Under RLM, multicast receivers adapt to both the static heterogeneity of link bandwidths as well as dynamic variations in network capacity (i.e., congestion). We describe the RLM protocol and evaluate its performance with a preliminary simulation study that characterizes user-perceived quality by assessing loss rates over multiple time scales. For the configurations we simulated, RLM results in good throughput with transient short-term loss rates on the order of a few percent and long-term loss rates on the order of one percent. Finally, we discuss our implementation of a software-based Internet video codec and its integration with RLM.

This paper presents the Reliable Multicast Protocol (RMP). RMP provides a totally ordered, reliable, atomic multicast service on top of an unreliable multicast datagram service such as IP Multicasting. RMP is fully and symmetrically distributed so that no site bears an undue portion of the communication load. RMP provides a wide range of guarantees, from unreliable delivery to totally ordered delivery, to K-resilient, majority resilient, and totally resilient atomic delivery. These QoS guarantees may be selected on a per packet basis. RMP provides many communication options, including virtual synchrony, a publisher/subscriber model of message delivery, a client/server model of delivery, an implicit naming service, mutually exclusive handlers for messages, and mutually exclusive locks.

This paper describes SRM (Scalable Reliable Multicast), a reliable multicast framework for application level framing and light-weight sessions. The algorithms of this framework are efficient, robust, and scale well to both very large networks and very large sessions. The framework has been prototyped in wb, a distributed whiteboard application, and has been extensively tested on a global scale with sessions ranging from a few to more than 1000 participants. The paper describes the principles that have guided our design, including the IP multicast group delivery model, an end-to-end, receiver-based model of reliability, and the application level framing protocol model. As with unicast communications, the performance of a reliable multicast delivery algorithm depends on the underlying topology and operational environment. We investigate that dependence via analysis and simulation, and demonstrate an adaptive algorithm that uses the results of previous loss recovery events to adapt the control parameters used for future loss recovery. With the adaptive algorithm, our reliable multicast delivery algorithm provides good performance over a wide range of underlying topologies.

We study the traffic that scalable multicast protocols generate in terms of message delays over the network as well as traffic counts at the link level in the case of self-similar sources. In particular, we study Bimodal Multicast and Scalable Reliable Multicast protocols proposed for scalable reliable multicasting. These protocols are based on different mechanisms for recovering from message losses and providing scalability. We discuss the protocol mechanisms as the main underlying factor in our empirical results. Our results can be considered as a contribution to the general problem of integration of multicast communication to large scale.

We present a new reliable multicast protocol, called ARM for adaptive reliable multicast. Our protocol integrates ARQ and FEC techniques. The objectives of ARM are (1) reduce the message overhead due to NACK requests, (2) reduce the amount of data transmission, and (3) reduce the time it takes for all receivers to receive the data intact (without loss). During data transmission, the sender periodically informs the receivers of the number of packets that are yet to be transmitted. Based on this information, each receiver predicts whether this amount is enough to recover its losses. Only if it is not enough, that the receiver requests the sender to encode additional redundant packets. Using ns simulations, we show the superiority of our hybrid ARQ-FEC protocol over the well-known scalable reliable multicast (SRM) protocol.