We investigate the problem of destination address resolution in closed multicast networks. By making use of the recirculating property of the closed networks, we propose a lookup node-based destination address resolution (LUNDAR) mechanism and a hybrid approach that facilitate destination address resolution by means of a lookup process. We show that these two strategies result in a tradeoff between the memory consumption and the incurred throughput reduction. Depending on the relative costs of memory and throughput, the network parameters can be adjusted to obtain the best balance between these costs.

In asynchronous transfer mode (ATM) switching systems, switch fabrics must be able to provide not only internally conflict-free paths for cells, but also quality of service (QoS) guarantee for calls. Most research concerning QoS guarantee focuses on the service disciplines of output-buffered switches. However, the scalability of output-buffered switches is rather low because of high implementation complexity. Previously, a new switching implementation called the cross-path switch has been proposed to achieve high throughput and low complexity. This paper addresses fundamental issues related to service disciplines in such a switch that are needed to achieve optimal performance. In particular we propose an algorithm to implement a multi-level stop-and-go queueing strategy that integrates the routing and scheduling functions. We show that our algorithm provides a tight delay bound and ensures that desirable traffic characteristics are preserved when cell traverse across the switch.

This paper investigates in detail the blocking and nonblocking behavior of multirate Clos switching networks at the connection/virtual connection level. The results are applicable to multirate circuit and fast-packet switching systems. Necessary and sufficient nonblocking conditions are derived analytically. Based on the results, an optimal bandwidth partitioning scheme is proposed to reduce switch complexity while maintaining the nonblocking property. The blocking behavior of blocking switches supporting multicast connections is investigated by means of simulation. We propose a novel simulation model that filters out external blocking events without distorting the bandwidth and fanout (for multicasting) distributions of connection requests. In this way, the internal blocking statistics that truly reflect the switch performance can be gathered and studied. Among many simulation results, we have shown that for point-to-multipoint connections, a heuristic routing policy that attempts to build a narrow multicast tree can have relatively low blocking probabilities compared with other routing policies. In addition, when small blocking probability can be tolerated, our results indicate that situations with many large-fanout connection requests do not necessarily require a switch architecture of higher complexity compared to that with only point-to-point requests.

Multicast capability can be incorporated into any interconnection network by using a general packet replication scheme. The network can then be used for both packet replication and routing processes. Unfortunately, such a multicast network can easily evolve to saturation due to instability. Once the network is saturated, the throughput drops to zero. The operation of the multicast network must therefore be carefully controlled to avoid the unstable region. Not well-thought-out control schemes, however, may result in a small throughput and inefficient utilization of the network. This paper investigates the stability issue in the multicast shuffle-exchange network in detail. Several schemes are then proposed to remove network instability. Our study indicates that a dynamic access control scheme can potentially achieve high network throughput even under nonuniform traffic conditions.

This paper examines the application of a simple yet general packet replication scheme to achieve multicasting in closed interconnection networks. The performance of these networks is studied and a general throughput equation is obtained to express the overall network throughput in terms of the average routing delay of the corresponding point-to-point network. The multicast performance results are thus built on top of the previously-established, and often simpler, point-to-point results. Making use of this formula, we investigate the multicast closed shuffle-exchange network in detail. Analytical results are compared with simulation results to obtain further insights into the network operation and ways to improve the network performance.