In this paper the authors present an overview of the existing parallel architecture types, followed by a brief survey of the algorithms which have been developed for power system simulation utilizing various parallel ...
详细信息
In this paper the authors present an overview of the existing parallel architecture types, followed by a brief survey of the algorithms which have been developed for power system simulation utilizing various parallel processors. The authors present their main contribution in the final section by presenting recent efforts in the application of waveform relaxation methods to power systems. The authors correlate the physical properties of the power system to the rate of convergence of the numerical waveform iterations.
An edge is a bisector of a simple path if it contains the middle point of the path. In this paper, efficient parallel algorithms are proposed on the EREW PRAM for the single-source and all-pairs tree bisector problems...
详细信息
ISBN:
(纸本)0769503500
An edge is a bisector of a simple path if it contains the middle point of the path. In this paper, efficient parallel algorithms are proposed on the EREW PRAM for the single-source and all-pairs tree bisector problems. Two O(log n) time single-source algorithms are proposed. One uses O(n) work and the other uses O(nlog n) work. The one using O(n) work is more efficient but only applicable to unweighted trees. One all-pairs parallel algorithm is proposed. It requires O(log n) time using O(n/sup 2/) work.
parallel algorithms on a mesh-connected computer with multiple broadcasting mechanism are presented. The solution times of these algorithms are superior to those obtained on a mesh without broadcasting or a mesh with ...
详细信息
parallel algorithms on a mesh-connected computer with multiple broadcasting mechanism are presented. The solution times of these algorithms are superior to those obtained on a mesh without broadcasting or a mesh with single global broadcasting. The results show that the time required to solve even simple problems can be reduced by careful analysis and design. It is assumed that a broadcast reaches all processors in a row or a column in constant time, t/sub b/. In practice, broadcasting may take t/sub b/log n time. In this case, the time complexity for the algorithms discussed becomes O(n/sup 1/ /sup 2/log n). Another extension is to allow multiple access to a broadcast bus; for example, all processing elements (PEs) can broadcast a bit simultaneously. Such schemes are physically realizable. Thus the sum of a row of PEs can be found in O(m) steps, where m is the length of the words. Therefore the time required for sorting n elements and selecting the kth largest number on an n*n mesh with such schemes can be further reduced.< >
The class of k-trees generalize the notion of trees, maximal outer-planar graphs, and caterpillars. We present new parallel algorithms to recognize k-trees and to find a collection of k vertex disjoint paths between a...
详细信息
The class of k-trees generalize the notion of trees, maximal outer-planar graphs, and caterpillars. We present new parallel algorithms to recognize k-trees and to find a collection of k vertex disjoint paths between a specified vertex pair (u,v), called a uv-cable, in a k-tree. A parallel algorithm to compute the k paths in a k-tree is also presented. The algorithms are based on parallel construction of a directed graph as a representation for R-trees, referred to as underlying trees in this paper. The model of parallel computation used is the CRCW PRAM (concurrent read, concurrent write, parallel RAM), where more than one processor can concurrently read from or write into the same memory location during the same memory cycle. Writing conflicts are resolved in a nondeterministic fashion. The recognition algorithm runs in O(log/sup 2/ n) time using O(m+n) processors. Constructing a spatial graph and the underlying tree representations of the k-tree for fixed k takes only O(log n) time using O(n) processors. The parallel algorithms for a uv-cable and k paths take O(log n) time using O(n) processors if the underlying tree structure of the k-tree is available.< >
In this paper we describe efficient parallel algorithms for computing canonical chains and canonical anti-chains partition of a set of points on a plane. The problem to compute chain and anti-chain partition is of int...
详细信息
In this paper we describe efficient parallel algorithms for computing canonical chains and canonical anti-chains partition of a set of points on a plane. The problem to compute chain and anti-chain partition is of interest in VLSI design, computational geometry and in computational molecular biology. A new affine transformation on the set of points is defined which transforms chains in the original point set into anti-chains in the transformed point set.
We examine the signal processing problem of space-time adaptive processing (STAP) from a computational point of view. Specifically, we implement an STAP algorithm on the IBM Powerparallel SP1 computer. The two main ch...
详细信息
We examine the signal processing problem of space-time adaptive processing (STAP) from a computational point of view. Specifically, we implement an STAP algorithm on the IBM Powerparallel SP1 computer. The two main challenges are solving multiple least squares problems simultaneously and finding an efficient way of implementing multiple computational steps that demand different optimal data distributions in parallel. The effectiveness of our implementation is demonstrated with experimental results.< >
The authors consider parallel algorithms for online (real-time) data compression that use dynamic (adaptive) learning by textual substitution. The author's model of computation is a systolic pipe, in which process...
详细信息
The authors consider parallel algorithms for online (real-time) data compression that use dynamic (adaptive) learning by textual substitution. The author's model of computation is a systolic pipe, in which processors are arranged in a linear array (each processor connected only to its left and right neighbors). He first considers an array of processors that is large enough to accommodate one dictionary entry per processor. He then discusses using an existing parallel machine where the number of available processors might be limited.< >
The authors present a simple parallel algorithm to height-balance a binary tree. The algorithm accepts any arbitrary binary tree as its input and yields an optimally shaped binary tree. For any arbitrary binary tree o...
详细信息
The authors present a simple parallel algorithm to height-balance a binary tree. The algorithm accepts any arbitrary binary tree as its input and yields an optimally shaped binary tree. For any arbitrary binary tree of n nodes the algorithm has a time complexity of O(lgn) and utilizes O(n) processors on a EREW PRAM model. The algorithm uses Euler tours and list ranking, which form the building blocks for many parallel algorithms.< >
A description is given of two algorithms that compute the Hough transform for straight lines on N*N images using 1*N processing arrays. The algorithms are developed for 1*N array processors and have been implemented o...
详细信息
A description is given of two algorithms that compute the Hough transform for straight lines on N*N images using 1*N processing arrays. The algorithms are developed for 1*N array processors and have been implemented on the AIS-5000 parallel vision computer. The complexity of both algorithms is O(N/sup 2/+PN) on the 1*N array processor (P is the number of angles polled which determines the theta -resolution of the Hough space), which compares well with the known optimal O(N+P) algorithms for N*N mesh arrays.< >
The author presents two parallel algorithms for the eigenvalue assignment problem along with details of implementations of these algorithms both on shared-memory and on distributed-memory parallel computers. A brief m...
详细信息
The author presents two parallel algorithms for the eigenvalue assignment problem along with details of implementations of these algorithms both on shared-memory and on distributed-memory parallel computers. A brief mention is also made of a parallel algorithm for the multi-input observer matrix equation. Some actual computational speed-up results of these algorithms on Alliant FX/8 and CRAY Y-MP are reported.< >
暂无评论