Data networks
TeraNet: a multi-gigabits per second ATM network
Computer Communications
A Fast Distributed Shortest Path Algorithm for a Class of Hierarchically Clustered Data Networks
IEEE Transactions on Computers
VLSI Signal Processing Technology
VLSI Signal Processing Technology
Bidirectional versus Unidirectional Networks: Cost/Performance Trade-Offs
MASCOTS '95 Proceedings of the 3rd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems
High performance wide-area overlay using deadlock-free routing
Proceedings of the 18th ACM international symposium on High performance distributed computing
Hi-index | 0.00 |
A performance metric called receptivity is introduced for quantifying the degree ofconcurrent communication possible in high-speed wide area networks (WAN's). Given astochastic demand pattern model, receptivity is defined to be the probability that allrequested connections can be established concurrently. Because calculation of the exactvalue of receptivity is shown to (generally) have an exponential complexity, an analyticestimate for its value is derived. The derived estimate is dependent on networkparameters such as the number of links, link capacity values, and a weighted hopdistance metric (which depends on the topological structure of the network and itsrelationship to parameter values of the stochastic model for the demand patterns). Thederived estimate for the proposed metric compares reasonably well with simulated valuesfor several asymmetric topological structures ranging from planar meshes to randomgraphs. The utility of the estimate is twofold. First, it can be computed quickly, i.e., inpolynomial time. Second, its simple analytic form provides the network architect withinsight into some of the inherent limitations and consequences associated with topological design choices for high-speed WAN's.