In this letter, we propose "Torus Ring", which is a modified version of 2-level hierarchical ring. The Torus Ring has the same complexity as the hierarchical rings, since the only difference is the way it connects the local rings. It has an advantage over the hierarchical ring when the destination of a packet is the adjacent local ring, especially to the backward direction. Although we assume that the destination of a network packet is uniformly distributed across the processing nodes, the average number of hops in Torus Ring is equal to that of the hierarchical ring. However, the performance gain of the Torus Ring is expected to increase, due to the spatial locality of the application programs in the real parallel programming environment. In the simulation results, latencies of the interconnection network are reduced by up to 19%, with moderate ring utilization ratios.

Topology of a network greatly affects the network performance. Depending on the purpose of a network, a specific topology may perform much better than any other topologies. Since the ad hoc networks are formed for a specific purpose, determining, and constructing the network topology based on the application requirements will enhance system performance. This paper proposes Bluetooth scatternet forming protocol in which the network topology is determined by three parameters. The parameters affecting the topology are the number of maximum slaves in a piconet, the number of maximum piconets that a gateway Bluetooth device can service, and the number of loops needed in the formed scatternet. These parameters can be read from a script file prior to the network formation. This process of reading the important parameters from the file would give users freedom in determining the network topology. The proposed protocol also includes a role negotiation process to accommodate different capabilities of the participating devices. The negotiation process of the protocol allows the resource-limited nodes to participate in the network. Different types of scatternet topologies like star, mesh, ring and line can be formed by specifying the parameters. This paper also discusses theoretical information necessary for calculating network topologies in detail. The protocol is verified with help of simulations, and implementations using commercially available Bluetooth devices. The detailed results are also presented in this paper.

In this paper we propose a parallel and distributed computation of genetic local search with irregular topology in distributed environments. The scheme we propose in this paper is implemented with a tree topology established on an irregular network where each computing element carries out genetic local search on its own chromosome set and communicates with its parent when the best solution of each generation is updated. We evaluate the proposed algorithm by a simulation system implemented on a PC-cluster. We test our algorithm on four types topologies: star, line, balanced binary tree and sided binary tree, and investigate the influence of communication topology and delay on the evolution process.

The trivalent Cayley graph TCn was introduced and investigated in [1],[2]. Though "the diameter" was presented in [2], unfortunately it was not the diameter but an upper bound of it. In this paper, a lower bound of the diameter dia(TCn) of the trivalent Cayley graph TCn is investigated and the formula dia(TCn) = 2n - 2 for n 3 is established.

In this paper we consider a parallel and distributed computation of genetic algorithms on loosely-coupled multiprocessor systems. Loosely-coupled ones are more suitable for massively parallel processing and also more easily VLSI implementation than tightly-coupled ones. However, communication overhead on parallel processing is more serious for loosely-coupled ones. We propose in this paper a parallel and distributed execution method of genetic algorithm on loosely-coupled multiprocessor systems of fixed network topologies in which each processor element carries out genetic operations on its own chromosome set and communicates with only the neighbors in order to save communication overhead. We evaluate the proposed method on the multiprocessor systems with ring, torus, and hypercube topologies for benchmark problem instances. From the results, we find that the ring topology is more suitable for the proposed parallel and distributed execution since variety of chromosomes in the ring is kept much more than that in the others. Moreover, we also propose a new network topology called cone which is a hierarchical connection of ring topologies. We show its effectiveness by experimental evaluation.

Management of control functions in large computer networks is a very difficult problem. One of the effective way to overcome the difficulty is to introduce hierarchical control structure (network cluster) in the management. When a fault occurred in the cluster, routing information at some nodes in the network must be updated in order to react the fault. However, the number of such nodes can be reduced by introducing ingenious topology into the cluster. This paper presents a fundamental discussion on network topology for a network cluster. First, L-FT is defined to represent a degree of fault-tolerance in a cluster with respect to link failures. Secondly, the minimum link problem M is defined to find the minimum number of links to make the cluster L-FT. The following results are obtained. (1) For a network cluster with the fault-tolerant topology 1-FT, at least 2n-2 links have to exist in the cluster where n is the number of border nodes in the cluster. (2) As far as connectivity of the whole network is held, for multiple L link failures in a L-FT cluster, the update of routing information at each node is localized within only the cluster containing the failed links. (3) Several hierarchical networks with fault-tolerant conditions are presented as case studies for a LAN and a MAN.