Showing papers on "Network topology published in 1982"

Message routing in a computer network

Abstract: An adaptive distributed message routing algorithm that may be implemented in a computer program to control the routing of data messages in a packet message switching digital computer network. Network topology information is exchanged only between neighbor nodes in the form of minimum spanning trees, referred to as exclusionary trees. An exclusionary tree is formed by excluding the neighbor node and its links from the tree. From the set of exclusionary trees received a route table and transmitted exclusionary trees are constructed.

Geometric critical point analysis of lossless power system models

Abstract: Electric utility analysts today face an increasingly difficult task of formulating both long and short term operating plans which will provide at the same time efficient and economical operation while delivering reliable and uninterrupted service to electricity users. One of the key ingredients in this planning is a set of large scale simulations of the steady-state network performance under various anticipated operating conditions. Central to these analyses are the classical "load flow" equations which are the equilibrium equations for the "swing equations" which are a physically based model of the dynamic operation of an n -node power system. Despite the long standing and widespread use of these equations, there remain a number of very basic open questions: What are the number and nature of the equilibria of the swing equations? How many stable equilibrium operating points are there in an n -node electric power grid? In this paper some powerful analytical tools from topology and geometry are used to answer certain of these questions. It is well documented that the load flow equations comprise a formidable large scale system but what is interesting, and perhaps surprising, is that even for a small number of buses, these equations possess a rather rich and intricate qualitative behavior which has heretofor been only partially understood. Indeed, until now there was no complete statement in the literature concerning the number of load flows in a general three-bus network. In Theorem 2.7, we state that for the "generic" three-bus network there are, for sufficiently small power injections, either four or six real load flows and that, in either case, exactly one of those load flows is stable. This is a special case of the results we derive for a general n -bus powergrid. Our method consists in first transforming the load flow equations for a lossless electric power network by trigonometric substitutions into algebraic equations. This makes it possible to apply some deep and powerful results from algebraic geometry and intersection theory to study these equations. An obvious tool for determining the number of solutions is provided by the classical theorem of Bezout, but it is shown that for systems describing an n -machine network with n \geq 4 , this result cannot be directly applied because the solutions contain solution components of positive dimension "at infinity." A major result in this paper is a modified Bezout technique which allows us to compute the number of complex (and a fortiori an upper bound on the number of real) solutions to the load flow equations. Combining this with the classical Morse inequalities, we obtain very explicit results regarding the number of stable load flows for a given network topology and set of power injections. The cases of three and four machine networks are considered in detail.

Performance evaluation of data communication systems

Abstract: This paper is a tutorial and a survey of analytical methods in the evaluation of data communication networks. The major mathematical methods are Markov chains applied to discrete time systems and queueing theory. Emphasis is placed on the applications of the mathematical tools. The discussion follows the framework of the layered architecture. In the section on data link control, rigorous as well as "engineering" approaches are highlighted. In this area, models of great accuracy have been developed. In the path-control or routing layer, the major model is provided by Kleinrock's delay analysis of packet networks. Finite buffer pools still pose many problems. The method of the homogeneous network is introduced, a method which reduces complexity originating from the network topology in favor of more realistic protocol features. This thought is expanded into the layer of end-to-end protocols where the tandem-queue model is introduced and its application to the flow-control problem discussed.

A Methodology for the Design of Communication Networks and the Distribution of Data in Distributed Supercomputer Systems

Abstract: In this paper the combined problem of communication network design and file allocation is studied for supercomputer networks. The network topology is restricted to be maximally connected and of minimal diameter in order to enhance network reliability as well as to reduce communication cost and delay.

The Radial Sweep Algorithm for Constructing Triangulated Irregular Networks

Abstract: For some applications, triangulated irregular networks are superior to rectangular grids in modeling complex surfaces. The radial sweep algorithm is an improvement of TIN techniques.

Efficient schemes for parallel communication

Abstract: A fundamental problem in the theory of parallel computation is to find an efficient interconnection pattern between N processors that minimizes the number of lines entering or leaving each processor while enabling fast communication between the processors.A family of Balanced communication schemes for connecting N processors with only a constant number of lines entering or leaving each processor is defined. It is proved that this network topology enables a fully distributed algorithm to route in parallel N packets each located in distinct processors to their distinct destinations in O(log2N) steps.Thus we give an optimal solution to the above problem.

On the Validation and Analysis of a New Method for Power Network Connecticity Determination

Abstract: The analysis method and results of a power network connectivity determination algorithm, part of the CONCE project, the central part of ENHER's computerized dispatching system, are reported together with a validation of the latter through practical measurements. The algorithm is constructed by repeated invocation of the same connectivity determination procedure for an abstract graph. Although emphasis is given to a version consisting of two steps, substation and overall network, a three step imiplementation is also suggested together with its possible advantages. The main concepts upon which the algorithm is based are reviewed briefly. The analysis shows that the processor time involved in its execution is linearly proportional to the power network dimensions. This fact suggests that the algorithm could be used with advantage for large power networks.

On the Number of Permutations Performable by the Augmented Data Manipulator Network

Abstract: The Augmented Data Manipulator (ADM) network has been proposed as an interconnection network for SIMD machines. As one measure for comparing the ADM network to multistage cube networks, the number of distinct data permutations performable in a single pass through the ADM is examined. Techniques are given to count the number of settings of any stage of the network which are permutations. Upper and lower bounds on the number of distinct permutations performable by the network are proven.

A general class of processor interconnection strategies

Abstract: A new class of general topologies is proposed in this paper for interconnecting a large network of computers in parallel and distributed environment. These structures have been shown to possess small internode distances, fairly low number of links per node, easy message routing and large number of alternate paths that can be used in case of faults in the system. The interconnection is based on a mixed radix number system, presented in this paper. The technique results in a variety of structures for a given number of processors N, depending on the required diameter in the network.A bus oriented structure is also introduced here, based on the same mathematical framework. These structures possess only two I/O ports per processor and are also shown to have small internode distances.

Topologies associated with the compact open topology on H(U).

Abstract: Let H(U) be the space of holomorphic functions on an open subset U of a locally convex space. The authors study various topologies canonically associated with the topology of compact convergence on H(U), especially the bornological topology and the barrelled topology.

A generalized simulator for computer networks

Abstract: A generalized network simulator has been constructed to help investigate the performance of computer networks at a variety of levels. The simulator accounts for user-host, host-node (a node is a communications interface unit), and node-node interactions. The program is highly modular so that the user can change one level without affecting the others; this structure allows for innovation and efficient experimentation.Model validation has been a major consideration in the develop ment of the simulator. We have employed a formal procedure to validate models at the input-output level by comparing them to simple lumped models for a given experimental frame.The simulator can handle a variety of network structures and protocols. A particular network is described in terms of a schedul ing function, a grouping function, and a distance control matrix. Several network topologies and protocols have been simulated successfully.One problem with the modular approach is that it may lead to inefficient programs. Our simul...

Computational Techniques for an Electrical Distribution and Information Control System

Abstract: This paper discusses the network analysis functions that are associated with the Electrical Distribution and Information Control System (EDICS for the real-time monitoring and control of the North York Hydro electrical distribution system. The methods presented can be adapted to any distribution network topology and automated control philosophy. Dynamic color displays are provided to effectively show the electrical state of the network, while a unique analysis method estimates network flows in the presence of circuit loops and feeder parallels. A simulation study subsystem provides a means for the dispatcher to verify the consequences of switching orders before they are issued and to investigate potential overload conditions.

State-space approach for the analysis of networks containing lossy coupled transmission lines in inhomogeneous media

Abstract: A general and powerful technique for the analysis of networks containing lossy coupled transmission lines in inhomogeneous media is presented. The technique is suitable for computer-aided procedures and no restrictions on network topology are required. Good agreement was found between measured and calculated responses. One of the applications of the derived equations for lossy lines is the design of microstrip circuits on semiconductor substrates. The effect of varying the substrate resistivity has been investigated.

New Fiber Optic Data Bus Topology

Abstract: A fiber optic data bus topology is introduced which avoids both the "central failure point" problem that afflicts single star topologies and the intermessage dynamic range problem evident in systems using directional couplers.© (1982) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A Comparison of Cube Type and Data Manipulator Type Network

Abstract: : The interconnection of a large number of processors and other devices to form a parallel/distributed computing system is a research area receiving a great deal of attention. One method is to use a multistage network. This paper compares two classes of multistage networks by examining two representative networks: the Generalized Cube and the Augmented Data Manipulator. The two topologies are compared using a graph theoretic approach. By interpreting the graphical representations of the networks in different ways, different implementations result. The costs of the various implementations are compared taking very large scale integration considerations into account. Finally, the robustness of the different networks is measured and contrasted. (Author)

6 – network topology

Abstract: Publisher Summary This chapter discusses some topological network problems, the boundary and co-boundary operators in network topology and their application to two problems, one electrical, the other economic. The topology of networks revolves around two operators, the boundary and the co-boundary. The boundary ∂ transforms numerical functions of edges into numerical functions of vertices. The co-boundary δ changes functions of vertices into functions of edges. When the boundary is used, the functions of edges or vertices are called 1-chains or 0-chains, and they are represented by column vectors. The coordinates of these column vectors are the functional values at the various edges or vertices. In the co-boundary context, the functions are 1-cochains and 0-cochains, represented by row vectors. In a network, a path is simple if it never passes through a vertex more than once. At most two edges in a simple path can have any particular vertex as endpoint and no two edges can be equal or inverse to each other. The initial and terminal points of a path are the initial point of the first edge and the terminal point of the last. The path is closed if its initial and terminal points coincide.

Graph theoretic approach to multistage interconnection networks

Abstract: Several multiprocessor/multicomputer systems have reported the use of interconnection networks which are composed of several stages of 2-input/2-output Switching Elements, and are called Multistage Interconnection Networks (MINs). Most of the existing MINs are topologically equivalent to each other, thereby showing their functional equivalence. In other words, there exist equivalent permutations between two different MINs if inputs of the first or the outputs of the last stage are renamed. Naturally, it is interesting to investigate new kinds of topology as they may provide different permutation sets compared to the existing topology. Furthermore it seems that there exists a particular topology that may provide better performance or some special characteristics. The current analytical procedure of designing MIN does not help in understanding the network topology. But, altogether a new approach is needed which is more versatile, convenient, and powerful than the existing techniques of analyzing and designing MIN. The graph theoretic analysis presented here provides an effective way of incorporating any desired type of topology in the resulting MIN. We begin our study by applying this analytical tool to the existing MINs to obtain a graph model of the networks. Thereafter, partitioning of the network graph into several bipartite subgraphs, we characterize the topology with the type and the number of the loops in each subgraph. The reverse process enables us to design all possible non-equivalent MINs and this leads to a general construction algorithm for the MINs. One special topology of MINs, called a Zeta ((zeta)) network, is developed and investigated for possible permissible permutations. The usefulness of this network topology is demonstrated in implementing a new Generalizer with a new efficient control algorithm that provides much improved set-up time over existing algorithms. Three-state cells are also used in designing easily testable MIN so that any one test could detect single or multiple stuck type faults. In addition, we introduce various groups of single-staged MINs for easy VLSI implementation of the MIN for large number of inputs.

Design Of A Passive Fiber Optic Data Bus For Shipboard Application

Abstract: Under contract from the Canadian Department of National Defence, Sperry Univac and Canstar Communications are developing a 10-megabit data link for military shipboard application. The system is based on a passive optical fiber network connecting 32 electro-optic bus access modules, each module being capable of servicing eight bus users. The key aspect of the project was to determine the optimum network topology based on considerations of optical loss, system security, simplicity of design, and replaceability of key components, and to demonstrate the capability of achieving the requirements with existing technology in electro-optic transmitters and receivers and in fiber optic access couplers. The proposed paper will discuss the overall system requirements, the selection of the optimum passive topology and the trade-offs associated with the alternatives, as well as outline the measured performance of a prototype system resulting from the initial phase of development.© (1982) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Network computer operating systems and task force scheduling

Abstract: The subject of this dissertation is network computer operating systems. It is argued that operating system techniques intended for use on network computers (modular computers, multi-microprocessors) should not depend on physical connection topology, much as conventional techniques for uniprocessors generally do not depend on particular machine architectures. To this end, a hierarchical control structure schema is introduced. The schema abstracts away from the physical topology of a network in favor of a tree-like logical framework. The control hierarchy forms a superstructure around which operating system techniques for processor scheduling, deadlock and failure recovery, and resource allocation can be built independently of physical topology. An algorithm for establishing efficient hierarchies in network computers is introducted and elevated with computer simulations. The scheduling of task forces in a network computer is taken up next. A distributed assignment protocol, named Wave Scheduling, which uses the control hierarchy for assigning network processors to task forces is presented and then analyzed by means of a cost model. The relative efficiency of the technique is shown to compare very favorably with respect to a central scheduler in networks with reasonable workloads. Finally, the MICROS network computer nodal operating system is discussed. MICROS is composed of a packet switching subsystem and a host processor control segment. It is written entirely in Concurrent Pascal and executes in the nodes of the MICRONET multi-microprocessor, a re-configurable network of DEC LSI-11 microcomputers.

Network Synchronization Initialization Performance of a TDMA System in the Ground Environment

Abstract: Results from a digital computer simulation of a network of approximately 50 nodes, typifying a spread spectrum time division multiple access (TDMA) system such as the Joint Tactical Information Distribution System (JTIDS), in a ground environment, are presented, and discussed. The results apply to initialization performance of network synchronization. The results provide an insight into network management tradeoffs which may be made with respect to: a) method of network synchronization, b) system capacity needed for network synchronization, and c) degree of network connectivity. The results provide qualitative as well as quantitative assessment of the impact of changing any of the above network management variables when the others remain fixed at various operating points. Since a wide range of time delays, typically from, few to many minutes may be incurred, it is concluded that network management techniques may be optimized to meet the initialization performance requirements of a given network. Such optimizations, however, depend upon specific implementations of the network synchronization process, the dynamics of the network nodes and the environment (electromagnetic, terrain, and weather) affecting their connectivity.

Computer Modeling of the MX Simulcast Radio Network

Abstract: Nodes of the MX system can communicate effectively in a trans-or post-attack environment by using a simulcast network that employs groundwave propagation For a sufficiently dense original network, the simulcast approach is relatively insensitive to degrading channel conditions and loss of nodes and links Network performance is a complicated function of the radio and antenna designs, terrain in the deployment area, destruction level, network topology and method of control, and channel conditions In order to systematically evaluate the key parameters affecting performance and, in turn, to optimize their joint selection for various deployment alternatives, the radio network simulator (RNS) was developed This paper focuses on the application of the RNS to the system design process After a discussion of the modeling techniques used in the simulation, its utilization for performance assessment and tradeoff evaluation are presented Specific areas considered include data rate and attack analysis

Design of Survivable Circuit-Switched Communication Networks

Abstract: We address the problem of designing a circuit-switched network for voice communications operating in a military environment. The circuit-switched network design problem may be briefly stated as: given the topology, route tables and control discipline, end-to-end offered traffic and performance requirements, determine trunk group sizes such that the requirements are satisfied. One of the key requirements of a design is that the network be survivable, where survivability is based on different destruction scenario conditions. Our objective is to guarantee an acceptable level of performance for every node pair and under each of the different anticipated damage scenarios. The main contributions of our present work are the development of approaches for designing networks that simultaneously satisfy performance requirements for different destruction scenarios. We describe the architecture of a survivable, circuit-switched network. The key characteristics of the survivable network design problem are highlighted and differences with respect to classical trunk sizing problem are pointed out. An important aspect of our work is that the sizing is based on the logical topology of the network rather than its trunk group topology. One design approach, which we have used successfully, is presented in detail.