Feedback Implosion Research Articles

A router-assisted session tree configuration mechanism for reliable multicast

A session tree based mechanism provides an efficient method to avoid well-known feedback implosion. However, it is not easy to configure an efficient session tree for IP multicast because it does not provide any explicit membership and routing topology information to the upper layer protocol. Incongruity between a session tree built on the transport layer and the corresponding routing tree on the network layer would incur large cost to handle control messages. This problem can be solved if a router that knows the information of routing topology can support the configuration of a session tree. Thus this letter proposed a router-assistant mechanism which minimizes the change of router functions and allows the routers to assist in providing a reliable multicast transport service.

On the scaling of feedback algorithms for very large multicast groups

Feedback from multicast group members is vital for many multicast protocols. In order to avoid feedback implosion in very large groups feedback algorithms with well-behaved scaling-properties must be chosen. In this paper, we analyse the performance of three typical feedback algorithms described in the literature. Apart from the basic trade-off between feedback latency and response duplicates we especially focus on the algorithms’ sensitivity to the quality of the group size estimation. Based on this analysis we propose a generalised framework for feedback algorithms and especially give recommendations for the choice of well-behaved feedback mechanisms that are suitable for very large groups.

A study of proactive hybrid FEC/ARQ and scalable feedback techniques for reliable, real-time multicast

In this paper, we examine the performance benefits of using parity encoding for reliable delivery of data to multiple receivers, where the receivers have heterogeneous loss and delay characteristics. First, we examine the effect of sending parity loss repairs proactively to receivers before it is known whether or not the repairs are needed to repair losses. We show that in many cases, proactive repair can reduce feedback implosion and the expected delay of reliable delivery without increasing bandwidth usage. Next, we examine how proactive repairs can be used to meet end-to-end deadlines to within a tunable probability. Last, we examine several receiver feedback mechanisms for parity repairs in a heterogeneous networking environment, and find that the various mechanisms offer different levels of robustness. Our examinations take into account temporally correlated (bursty) loss, as well as loss of feedback messages from receivers.

Selection of representatives for feedback suppression in reliable multicast protocols

The representative scheme in which a small set of multicast group members are selected as representatives to represent congested areas and hence suppress feedback from other receivers is an effective way of avoiding feedback implosion. Two algorithms used for selecting group members as representatives are further investigated. It is shown through simulation that the algorithm that selects representatives using the loss probability observed at the receivers is more effective in reducing the feedback returned.

Adaptive approach for FEC reliable multicast

An adaptive integrated forward error correction approach to provide reliability in a multicast environment is proposed. The proposed solution avoids the undesirable feedback implosion phenomenon regardless of the number of receivers and uses minimum bandwidth in the feedback channel.

Scalable fair reliable multicast using active services

Scalability is of paramount importance in the design of reliable multicast transport protocols, and requires careful consideration of a number of problems such as feedback implosion, retransmission scoping, distributed loss recovery, and congestion control. In this article, we present a reliable multicast architecture that invokes active services at strategic locations inside the network to comprehensively address these challenges. Active services provide the ability to quickly and efficiently recover from loss at the point of loss. They also exploit the physical hierarchy for feedback aggregation and effective retransmission scoping with minimal router support. We present two protocols, one for packet loss recovery and another for congestion control, and describe an experimental testbed where these have been implemented. Analytical and experimental results are used to demonstrate that the active services architecture improves resource usage, reduces latency for loss recovery, and provides TCP-friendly congestion control.

Scalable feedback for large groups

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 works well in case of homogeneous and heterogeneous delays. We apply the feedback mechanism to reliable multicast and compare it to existing timer-based feedback schemes. Our mechanism achieves lower negative acknowledgment character (NAK) latency for the same performance in terms of NAK suppression. No topological information of the network is used, and data delivery is the only support required from the network. The mechanism adapts to a dynamic number of receivers and leads to a stable performance for implosion avoidance and feedback latency.

Scalable and reliable ATM multicast employing RM cell consolidation

The authors present a new scalable and reliable ATM multicast algorithm (SRAM) that can provide reliable data transfer services. This algorithm employs the resource management cell consolidation mechanism for available bit rate point-to-multipoint connections which resolves the feedback implosion problem without increasing the number of connections according to the number of receivers. The proposed scheme is compared with other reliable multicast schemes through numerical analysis under reasonable assumptions.

Optimal deterministic timeouts for reliable scalable multicast

Reliable multicast protocols suffer from the problem of feedback implosion. To avoid this problem, the number of receivers sending feedback in case of loss must be small. However, losses experienced by different receivers are strongly correlated, since receivers share common resources in the multicast tree. One approach to feedback implosion avoidance relies on delaying feedback at the receivers. We present deterministic timeouts for reliable multicast (DTRM), a distributed algorithm to compute optimal deterministic timeouts for each receiver in a multicast tree as a function of the tree topology and the sender-to-receiver round-trip delays. DTRM has several desirable properties. First, feedback implosion is provably avoided for a single loss anywhere in the tree, provided delay jitter is bounded. Second, the computation of the timeouts can be entirely distributed; receivers and intermediate nodes only rely on local topology information. Third, the timeouts computed by DTRM are optimal with respect to the maximum response time.