Showing papers in "Journal of Parallel and Distributed Computing in 1989"
••
TL;DR: This paper completely analyze the hypercube network by explicitly computing the eigenstructure of its node adjacency matrix and shows that a diffusion approach to load balancing on a hypercube multiprocessor is inferior to another approach which is called the dimension exchange method.
1,074 citations
••
TL;DR: It is shown that, by adding 20% silicon area over each processor, connection autonomy can deliver orders of magnitude of performance improvement over the same network without connection autonomy.
148 citations
••
TL;DR: A unified formulation is provided for both the retrieving and the learning phases of most ANNs, and on the basis of the formulation, a programmable ring systolic array is developed that maximizes the strength of VLSI in terms of intensive and pipelined computing and yet circumvents the limitation on communication.
127 citations
••
TL;DR: A distributed processor/memory organization is developed for efficiently simulating asynchronous, value-passing connectionist models and the structural model offers an efficient methodology for mapping virtual neural networks onto a real parallel computer.
89 citations
••
TL;DR: It is shown that the greedy algorithm finds provably near-optimal schedules in polynomial time and that optimal schedules can be found in polynnomial time when the number of distinct types of workstations is bounded by a constant.
68 citations
••
TL;DR: Issues related to the design of pseudorandom number generators (prn's) for MIMD (multiple instruction stream/multiple data stream) parallel processors, which are very well suited to Monte Carlo calculations are discussed and analyzed.
68 citations
••
TL;DR: It is shown for a number of combinatorial optimization problems how they can be mapped directly onto a Boltzmann machine by choosing appropriate connection patterns and connection strengths, equivalent to finding an optimal solution of the corresponding optimization problem.
66 citations
••
TL;DR: This paper describes a new divide-and-conquer algorithm that solves both problems and investigates the relationship between them, an important method of algorithm design: detailed combinatorial analysis of an overconstrained variation can reveal solutions to the corresponding original problem.
61 citations
••
TL;DR: This paper presents a method for determining the virtual states that describe the reachable program states, and the reduction in the number of states is analyzed.
61 citations
••
TL;DR: Two new and efficient parallel algorithms which detect lines by performing a Hough transform are presented, which perform only simple data movement operations over relatively short distances.
58 citations
••
TL;DR: A parallel αβ searching program that achieves high performance through the use of four different types of processes: Controllers, Searchers, Table Managers, and Scouts is described.
••
TL;DR: This work defines multilevel inclusion properties, gives some necessary and sufficient conditions for these properties to hold in multiprocessor environments, and shows their importance in reducing the complexities of cache coherence protocols.
••
TL;DR: In this article, the authors present a conceptual model that explores the potential of artificial intelligence tools, specifically expert systems, to design an Intelligent Operating System for multiprocessor systems.
••
TL;DR: A novel modeling methodology is described for evaluating the performance of distributed, multiple-computer systems that employs a set of analytic tools to obtain an estimate of the average execution time of a parallel implementation of a program (or transaction) in a distributed environment.
••
TL;DR: The algorithm presented in this paper embeds an N node binary tree into a hypercube with expansion O(log N) and unit dilation.
••
TL;DR: Analysis of the circuit reveals that it can proceed from an unbiased starting state to a stable state which can be interpreted as a satisfying truth value assignment for the 3-Satisfiability problem, if such an assignment exists.
••
TL;DR: The proposed implementation of distributed processing to approximating the solution to certain NP-complete problems in Combinatorial Optimization consists of a stochastic search inspired in the principles of Statistical Physics and a wide class of problems for which the method can be parallelized is characterized.
••
TL;DR: This paper uses a scheme which represents a quadtree by its leaf nodes; each node is represented by the coordinates of the upper left corner of its corresponding block and its color and size.
••
TL;DR: SDEF is intended to provide systolic algorithm researchers/ developers with an executable notation, and the software systems community with a target notation for the development of higher-level syStolic software tools.
••
TL;DR: An associative parallel processing system using novel VLSI associative memory is described, which proves to be a general-purpose parallel processor with a broad range of applications.
••
TL;DR: The present paper emphasizes the “domain-specific” structure of neural networks, as well as emphasizing that technologists have much to learn from the study of neurobiological systems.
••
TL;DR: It is proved that the Orthogonal Access Multiprocessing system exhibits the same performance as any other system with the same number of processors within a factor of 3, and is a strong competitor in the world of special-purpose parallel systems.
••
TL;DR: A "pipelined" scheme for performing k prefix operations in O(k + log p) time on p processors, which implies a similar pipelining result for the "data distribution" operation of Ullman [16].
••
TL;DR: The technique can be used to transform a class of algorithms to specific forms that can be mapped directly onto higher-dimensional systolic networks, yet maintaining the same number of processing cells as its 1-D counterpart.
••
TL;DR: In this article, an analytic approximation for the blocking probability of a circuit-switched, multipath multistage interconnection network when it is required to perform arbitrary permutations under local control is presented.
••
TL;DR: The unidirectional flow of control and data in the design assures implementation of the linear array in the known fault models of Wafer Scale Integration.
••
TL;DR: It is shown that the optimal balancing for acyclic connected data flow graphs generated from a data flow language can be formulated into certain linear programming problems which have efficient algorithmic solutions.
••
TL;DR: The possibility of using neural network optimization to perform optimized code generation, and possible benefits, are discussed, and a simple general method for constructing complicated syntactic constraints is introduced.
••
TL;DR: The proposed hierarchical approach leads to one-dimensional computational arrays for an important class of algorithms which need O(n3) computational steps in a sequential implementation.
••
TL;DR: A hierarchical Artificial Neuronal Network is proposed as a model sensorium wherein feedback is allowed to modify the categorization abilities of the system, yielding a more concrete representation.