Link Models for Networks with Dynamic Topologies

Abstract

Dynamic hierarchical networks represent an architectural strategy for employing adaptive behavior in applications sensitive to highly variable external demands or uncertain internal conditions. The characteristics of such architectures are described, and the significance of adaptive capability is discussed. The necessity for assessing the tradeoffs between performance improvements (reduced and bounded message transmission time, increased throughput) and the added costs (reconfiguration delays, redundant links, etc.) leads to the use of complex queueing models. The assumptions underlying the general model are stated, and a class of applicable models (queues in random environments or RE-queues) is introduced. Matrix-geometric methods are reviewed in terms of their suitability for addressing several variations of a subclass of RE-queue models. Matrix-geometric techniques are considered to offer the greatest promise for obtaining usable results for assessing the cost/benefit tradeoffs.