Jörg Nonnenmacher

Parity-based loss recovery for reliable multicast transmission

We investigate how FEC (Forward Error Correction) can be combined with ARQ (Automatic Repeat Request) to achieve scalable reliable multicast transmission. We consider the two scenarios where FEC is introduced as a transparent layer underneath a reliable multicast layer that uses ARQ, and where FEC and ARQ are both integrated into a single layer that uses the retransmission of parity data to recover from the loss of original data packets.To evaluate the performance improvements due to FEC, we consider different types of loss behaviors (spatially or temporally correlated loss, homogeneous or heterogeneous loss) and loss rates for up to 106 receivers. Our results show that introducing FEC as a layer below ARQ can improve multicast transmission efficiency and scalability and that there are substantial additional improvements when the two are integrated.

Optimal multicast feedback

We investigate the scalability of feedback in multicast communication and propose a new method of probabilistic feedback based on exponentially distributed timers. By analysis and simulation for up to 10/sup 6/ receivers we show that feedback implosion is avoided while feedback latency is low. The mechanism is robust against the loss of feedback messages and robust against homogeneous and heterogeneous delays. We apply the feedback mechanism to reliable multicasting and compare it to existing timer-based feedback schemes. Our mechanism achieves lower NAK (loss signaling) latency for the same performance in terms of NAK suppression. It is scalable, the amount of state at every group member is independent of the number of receivers. No topological information of the network is used and data delivery is the only support required from the network. It adapts to the number of receivers and leads therefore to a constant performance for implosion avoidance and feedback latency.

How bad is reliable multicast without local recovery

We examine the impact of the loss recovery mechanisms on the performance of a reliable multicast protocol. Approaches to reliable multicast can be divided into two major classes: source-based recovery, and distributed recovery. For both classes we consider the state of the art: for source-based recovery, a type 2 hybrid ARQ scheme with parity retransmission; for distributed recovery, a scheme with local multicast retransmission and local feedback processing. We further show the benefits of combining the two approaches and consider a type 2 hybrid ARQ scheme with local retransmission. The schemes are compared for up to 10/sup 6/ receivers under different loss scenarios with respect to network bandwidth usage and completion time of a reliable transfer. We show that the protocol based on local retransmissions via type 2 hybrid ARQ performs best for bandwidth and latency. For networks, where local retransmission is not possible, we show that a protocol based on type 2 hybrid ARQ comes close to the performance of a protocol with local retransmissions.

Bandwidth-allocation policies for unicast and multicast flows

Using multicast delivery to multiple receivers reduces the aggregate bandwidth required from the network compared to using unicast delivery to each receiver. However, multicast is not yet widely deployed in the Internet. One reason is the lack of incentive to use multicast delivery. To encourage the use of multicast delivery, we define a new bandwidth-allocation policy, called LogRD, taking into account the number of downstream receivers. This policy gives more bandwidth to a multicast flow as compared to a unicast flow that shares the same bottleneck, without starving the unicast flows, however. The LogRD policy also provides an answer to the question on how to treat a multicast flow compared to a unicast flow sharing the same bottleneck. We investigate three bandwidth-allocation policies for multicast flows and evaluate their impact on both receiver satisfaction and fairness using a simple analytical study and a comprehensive set of simulations. The policy that allocates the available bandwidth as a logarithmic function of the number of receivers downstream of the bottleneck achieves the best tradeoff between receiver satisfaction and fairness.

Performance comparison of centralized versus distributed error recovery for reliable multicast

We examine the impact of the loss recovery mechanism on the performance of a reliable multicast protocol. Approaches for loss recovery in reliable multicast can be divided into two major classes: centralized (source-based) recovery and distributed recovery. For both classes we consider the state of the art: for centralized recovery, an integrated transport layer scheme using parity multicast for error recovery (hybrid ARQ type 2) as well as timer-based feedback suppression, and for distributed recovery, a scheme with local data multicast retransmission and feedback processing in a local neighborhood. We also evaluate the benefits of combining the two approaches into distributed error recovery (DER) with local retransmissions using a type 2 hybrid ARQ scheme. The schemes are evaluated for up to 10/sup 6/ receivers under different loss scenarios with respect to network bandwidth usage and completion time of a reliable transfer. We show that using DER with type 2 hybrid ARQ gives best performance in terms of bandwidth and latency. For networks, where local retransmission is not possible, we show that a centralized protocol based on type 2 hybrid ARQ comes close to the performance of a protocol with local retransmissions.

Reliable multicast: where to use FEX

The MBONE provides an infrastructure for multicast communication on the internet based on IP. Several proposals have been made to create reliable multicast transport on top of the MBONE structure. Nearly all research on reliable multicast protocols for the MBONE focuses on ARQ error recovery. The counterpart of ARQ, FEC, does not guarantee 100% reliability but increases the reliability. The aim of this paper is to determine the best place in a multicast tree for the use of FEC. We develop a framework that allows us to model analytically the impact of FEC on the average number of transmissions necessary to transmit a packet to all members of the multicast group. We look on different multicast tree topologies, different degrees of correlated loss in the multicast tree, different multicast group sizes and show the effect of FEC in terms of transmissions needed to achieve 100% reliability. We find that the shared part of the multicast tree is not always the best part to employ FEC.

WAVE: A New Multicast Routing Algorithm for Static and Dynamic Multicast Groups

We present a new multicast algorithm called WAVE for establishing source-specific multicast trees. WAVE meets multiple quality of service requirements (constraints) such as delay, cost, and available bandwidth, simultaneously. Simulation results show that WAVE performs very good in terms of delay and cost for both, static and dynamic multicast groups, when compared to the best multicast algorithms known.

Reliable multicast via satellite : uni-directional vs. bi-directional communication

We investigate reliable multicast communication over satellite networks. We compare a scenario where receivers can use a feedback channel to signal loss to a scenario where no feedback channel is available. We show that the introduction of a feedback channel is the key to allow for bandwidth-efficient, robust and fully reliable multicast communication via satellite.

Research: The impact of routing on multicast error recovery

We investigate the relationship between multicast routing algorithms and reliable multicast communication. To capture the impact of the multicast tree topology onto reliable multicast, we consider two performance measures, namely the probability mass function of successful receptions and the expected number of retransmissions needed to deliver a packet successfully from the source to all receivers. Since the expected number of retransmissions is computationally expensive we also give a tight approximation. We finally evaluate the impact of routing algorithms on the performance of reliable multicast transmission and propose a realistic generic model for a multicast tree.

Center placement algorithms for large multicast groups

An increasing number of distributed applications require a specific form of multicast called dissemination, in which a single source reliably transfers data to multiple receivers. Reliability leads for large groups of receivers (100s or 1000s of participants) to the problem of feedback implosion at the source and to a decrease of transmission efficiency. The cluster approach was identified to have excellent scalability with the number of receivers. It partitions the multicast delivery tree into clusters, where a representative in the cluster called center is used for local feedback processing and local transmission. Up to now, clustering/center placement has been done administratively or based on network addresses. Needed are center placement algorithms, allowing the introduction of placement criteria based on the network topology and on delay. In this work, three center placement algorithms designed for static multicast groups are presented and simulation results are shown in order to asses their performance.