检索结果-内蒙古大学图书馆

IEEE International Conference on Intelligent Processing Systems

作者： Dai, WZ Hangzhou Inst Commerce Dept Elect Hangzhou 310035 Peoples R China

ISBN: (纸本)0780342534

In this paper, first, a fault diagnosis approach of large system based on iteration of function space in dynamic programming is proposed, and it new concept about the minimum of coefficient product of routes from source node to target node is put forward;an algorithm to figure out this minimum and to track down this correspondent route is given. Second, the algorithm is changed into a tabular iteration method, which is simpler, more regular and helpful for programming. Third, the algorithm's operation time is discussed. In order to reduce the operation time, a concept of bipartite state space is brought up. According to this concept, a large net can be divided into two or more subnets which are independent of each other. Fourth, a hind of parallel algorithm based on tabular iteration for fault diagnosis is also offered;the steps of applying this algorithm are detailed. Finally, this approach is shown to be more effective and simpler by an example of fault diagnosis.

关键词： fault diagnosis parallel algorithms dynamic programming minimisation graph theory parallel algorithm tabular iteration fault diagnosis function space iteration dynamic programming minimum coefficient product programming bipartite state space operation time shortest route

来源：评论

学校读者我要写书评

暂无评论

Minimizing communication penalty of triangular solvers by runtime mesh configuration and workload redistribution

引用

JOURNAL OF SUPERCOMPUTING 1999年第1期14卷 77-95页

作者： Wang, DQ Chu, E Univ Guelph Dept Math & Stat Guelph ON N1G 2W1 Canada

In this article, we study the effects of network topology and load balancing on the performance of a new parallel algorithm for solving triangular systems of linear equations on distributed-memory message-passing multiprocessors. The proposed algorithm employs novel runtime data mapping and workload redistribution methods on a communication network which is configured as a toroidal mesh. A fully parameterized theoretical model is used to predict communication behaviors of the proposed algorithm relevant to load balancing, and the analytical performance results correctly determine the optimal dimensions of the toroidal mesh, which vary with the problem size, the number of available processors, and the hardware parameters of the machine. Further enhancement to the proposed algorithm is then achieved through redistributing the arithmetic workload at runtime. Our FORTRAN implementation of the proposed algorithm as well as its enhanced version has been tested on an Intel iPSC/2 hypercube, and the same code is also suitable for executing the algorithm on the iPSC/860 hypercube and the Intel Paragon mesh multiprocessor. The actual timing results support our theoretical findings, and they both confirm the very significant impact a network topology chosen at runtime can have on the computational load distribution, the communication behaviors and the overall performance of parallel algorithms.

关键词： parallel algorithm triangular solver hypercube mesh torus overlap of communications by computations load balancing performance modeling and evaluation

来源：评论

学校读者我要写书评

暂无评论

A BSP Bareiss algorithm for Toeplitz systems

引用

JOURNAL OF parallel AND DISTRIBUTED COMPUTING 1999年第2期56卷 99-121页

作者： Huang, YG McColl, WF Univ Oxford Comp Lab Oxford OX1 3QD England

In this paper, a BSP (bulk synchronous parallel) Bareiss algorithm for Toeplitz system is described. We investigate various data distribution and scheduling strategies for mapping a typical class of systolic array algorithms onto BSP machines. Load balance, both in communication and computation, as well as linear speedup have been achieved for the Toeplitz system solver and at the same time the minimum memory requirement is achieved. An implementation has been tested on Sun workstations, an SGI Power Challenge, and an IBM SP?, using the Oxford BSPlib (Hill et al., 1997. (C) 1999 Academic Press.

关键词： bulk synchronous parallel Bareiss algorithm Toeplitz system parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

Fundamental data movement operations and its applications on a hyper-bus broadcast network

引用

parallel COMPUTING 1999年第2期25卷 137-157页

作者： Tsai, HR Horng, SJ Kao, TW Lee, SS Tsai, SS Natl Taiwan Univ Sci & Technol Dept Elect Engn Taipei Taiwan Overseas Chinese Coll Commerce Dept Informat Management Taichung Taiwan Kuang Wu Inst Technol & Commerce Dept Elect Engn Taipei Taiwan Fu Shin Inst Technol & Commerce Dept Elect Engn Ilain Taiwan

A hyper-bus broadcast network (HBBN) consists of processors only sharing by some global buses, and there are no local links between processors. Based on such an architecture, we will exploit several efficient time parallel algorithms for solving the well-known fundamental data movement problems which had been extensively studied by researchers and widely applied to the field of image processing, digitized geometry and computer graphics. These include the leftmost one problem, the prefix maxima/minima problem, the m-contour problem, the all nearest neighbor problem and the all nearest smaller values problem, respectively. Note that the proposed algorithms not only can be implemented on the HBBN but also can be easily modified to run on other broadcast-based networks with the same time and processor complexities. (C) 1999 Elsevier Science B.V. All rights reserved.

关键词： prefix maxima/minima computation m-contour problem all nearest neighbor problem all nearest smaller values problem image processing digitized geometry computer graphics parallel algorithm hyper-bus broadcast network

来源：评论

学校读者我要写书评

暂无评论

Design of optimal array processors for two-step division-free Gaussian elimination

引用

IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS 1999年第12期E82D卷 1503-1511页

作者： Peng, ST Sedukhin, SG Univ Aizu Aizuwakamatsu 9658580 Japan

The design of array processors for solving linear systems using two-step division-free Gaussian elimination method is considered. The two-step method can be used to improve the systems based on the one-step method in terms of numerical stability as well as the requirements for high-precision. In spite of the rather complicated computations needed at each iteration of the two-step method, we develop an innovative parallel algorithm whose data dependency graph meets the requirements for regularity and locality. Then we derive two-dimensional array processors by adopting a systematic approach to investigate the set of all admissible solutions and obtain the optimal array processors under linear time-space scheduling. The array processors is optimal in terms of the number of processing elements used.

关键词： linear system parallel algorithm parallel architecture systolic array processors

来源：评论

学校读者我要写书评

暂无评论

Mesh-nodal network analysis

引用

IEEE TRANSACTIONS ON POWER SYSTEMS 1999年第4期14卷 1375-1381页

作者： Dudurytch, I Gudym, V State Univ Lviv Polytech UA-290646 Lviv Ukraine

New topology-based hybrid method for electrical networks with magnetic couplings analysis is presented. Due to difficulties of general nodal formulation for the electrical circuits with magnetic couplings the network is split into two parts. First part contains all branches with magnetic couplings and its equations are formulated in mesh coordinates. The second part contains the rest of the network and its equations are formulated in nodal coordinates. Combination of two approaches for a single network with magnetic couplings yields new general mesh-nodal network analysis formulation with natural network decomposition into subsystems, suitable for parallel computational algorithm application. The example of proposed approach application is presented.

关键词： nodal analysis mesh analysis mesh-nodal analysis networks circuits magnetic couplings parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

Implementation of two-dimensional systolic algorithms for multi-layered neural networks

引用

JOURNAL OF SYSTEMS ARCHITECTURE 1999年第14期45卷 1209-1218页

作者： Song, Q Kho, KP See, KH Nanyang Technol Univ Sch Elect & Elect Engn Singapore 639798 Singapore

A new two-dimensional systolic algorithm is proposed in this paper for parallel implementation of the multi-layered neural network. To reduce communication overhead, the input data flow is passed along the horizontal and vertical directions of the systolic array alternately, over different layers of the neural network. This new algorithm accelerates learning process of the neural network. Transputer implementation of the proposed algorithm and experimental results are presented to show efficiency of the new algorithm. (C) 1999 Elsevier Science B.V. All rights reserved.

关键词： two-dimensional systolic array multi-layered neural network parallel algorithm transputer implementation

来源：评论

学校读者我要写书评

暂无评论

Computational geometry on the broadcast communication model

引用

JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 1999年第3期15卷 383-395页

作者： Yang, CB Natl Sun Yat Sen Univ Dept Appl Math Kaohsiung 804 Taiwan

In this paper, we solve three geometric problems, including the ranking, convex hull and closest pair problems, under the broadcast communication model. To solve these problems, we propose a general scheme, the p-division approach, which is based upon the divide-and-conquer strategy. In the 2-dimensional space, the time complexities of our algorithms for solving these problems are all O(n + n/p log n/p), where n is the number of input points and p is the number of processors used. Furthermore, our algorithms are all conflict-free and optimal. In the k-dimensional space, k greater than or equal to 3, our ranking algorithm requires O(n/p log(k-1) n/p n/p + n log(k-2) n/p) time.

关键词： broadcast communication computational geometry parallel algorithm divide-and-conquer conflict-free

来源：评论

学校读者我要写书评

暂无评论

On the complexity of hamiltonian path and cycle problems in certain classes of digraphs

引用

DISCRETE APPLIED MATHEMATICS 1999年第1-3期95卷 41-60页

作者： Bang-Jensen, J Gutin, G Odense Univ Dept Math & Comp Sci DK-5230 Odense Denmark Brunel Univ Dept Math & Stat Uxbridge UP8 3PH Middx England

We survey results on the sequential and parallel complexity of hamiltonian path and cycle problems in various classes of digraphs which generalize tournaments. We give detailed informations on the difference in difficulties for these problems for the various classes as well as prove new results on hamiltonian paths starting in a specified vertex for a quite general class of digraphs. (C) 1999 Elsevier Science B.V. All rights reserved.

关键词： hamiltonian path hamiltonian cycle polynomial algorithm parallel algorithm locally semicomplete digraphs path-mergeable digraph in-semicomplete digraph quasi-transitive digraph decomposable digraph multipartite tournament semicomplete multipartite digraph longest path hamiltonian connected path factor cycle factor flows path covering

来源：评论

学校读者我要写书评

暂无评论

Rearranging scattered information on BSR

引用

INFORMATION PROCESSING LETTERS 1999年第1期71卷 43-47页

作者： Xiang, LM Ushijima, K Kyushu Univ Dept Comp Sci & Commun Engn Higashi Ku Fukuoka 8128581 Japan

The problem of rearranging scattered information arises in many applications of data/information management systems. A simpler case of this problem is the so-called k-compaction problem, which requires Omega(root log n) EREW PRAM time, Omega (log log n) CREW PRAM time, or Theta (log k/ log log n) CRCW PRAM time using ii processors. In this paper, we give constant time parallel algorithms using n processors on BSR, another PRAM model, for the problem of rearranging scattered information and the k-compaction problem. (C) 1999 Published by Elsevier Science B.V. All rights reserved.

关键词： rearranging scattered information k-compaction problem BSR PRAM model parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：