检索结果-内蒙古大学图书馆

Techniques for compiling programs on distributed memory multicomputers

PARALLEL COMPUTING 1995年第12期21卷 1895-1923页

作者： Lee, PZ Institute of Information Science Academia Sinica Taipei Taiwan ROC

It is widely accepted that distributed memory parallel computers will play an important role in solving computation-intensive problems. However, the design of an algorithm in a distributed memory system is time-consuming and error-prone, because a programmer is forced to manage both parallelism and communication. In this paper, we present techniques for compiling programs on distributed memory parallel computers. We will study the storage management of data arrays and the execution schedule arrangement of Do-loop programs on distributed memory parallel computers. First, we introduce formulas for representing data distribution of specific data arrays across processors. Then, we define communication cost for some message-passing communication operations. Next, we derive a dynamic programming algorithm for data distribution. After that, we show how to improve the communication time by pipelining data, and illustrate how to use data-dependence information for pipelining data. Jacobi's iterative algorithm and the Gauss elimination algorithm for linear systems are used to illustrate our method. We also present experimental results on a 32-node nCUBE-2 computer.

关键词： data-dependence vector data partition and distribution distribution function distributed memory computer dynamic programming algorithm for data distribution parallelizing compiler pipelining

来源：评论

学校读者我要写书评

暂无评论

Efficient algorithms for data distribution on distributed memory parallel computers

引用

IEEE TRANSACTIONS ON PARALLEL AND distributed SYSTEMS 1997年第8期8卷 825-839页

作者： Lee, PZ Institute of Information Science Taipei Taiwan

Data distribution has been one of the most important research topics in parallelizing compilers for distributed memory parallel computers. Good data distribution schema should consider both the computation load balance and the communication overhead. In this paper, we show that data redistribution is necessary for executing a sequence of Do-loops if the communication cost due to performing this sequence of Do-loops is larger than a threshold value. Based on this observation, we can prune the searching space and derive efficient dynamic programming algorithms for determining effective data distribution schema to execute a sequence of Do-loops with a general structure. Experimental studies on a 32-node nCUBE-2 computer are also presented.

关键词： component alignment data distribution distributed memory computer Do-loops dynamic programming algorithm for data distribution parallelizing compiler

来源：评论

学校读者我要写书评

暂无评论

Parallelism in ILU-preconditioned GMRES

引用

PARALLEL COMPUTING 1998年第14期24卷 1927-1946页

作者： Vuik, C van Nooyen, RRP Wesseling, P Delft Univ Technol NL-2600 GA Delft Netherlands Delft Univ Technol Fac Civil Engn NL-2600 GA Delft Netherlands

A parallel implementation of the preconditioned GMRES method is described. The method is used to solve the discretized incompressible Navier-Stokes equations. A parallel implementation of the inner product is given, which appears to be scalable on a massively parallel computer. The most difficult part to parallelize is the ILU-preconditioner. We parallelize the preconditioner using ideas proposed by Bastian and Horton (P. Bastian, G. Horton, SIAM. J. Stat. Comput. 12 (1991) 1457-1470). Contrary to some other parallel methods, the required number of iterations is independent of the number of processors used. A model is presented to predict the efficiency of the method. Experiments are done on the Gray T3D, computing the solution of a two-dimensional incompressible flow. Predictions of: computing time show good correspondence with measurements. (C) 1998 Elsevier Science B.V. All rights reserved.

关键词： parallel ILU preconditioner distributed memory computer parallel performance model computational fluid dynamics

来源：评论

学校读者我要写书评

暂无评论

OPTIMIZATION OF ARRAY REDISTRIBUTION FOR distributed-memory MULTIcomputerS

引用

PARALLEL COMPUTING 1995年第9期21卷 1485-1490页

作者： WAKATANI, A WOLFE, M OREGON GRAD INST DEPT COMP SCI & ENGNPORTLANDOR 97291 MATSUSHITA ELECT IND CO LTD MORIGUCHIOSAKA 570JAPAN

Languages such as High Performance Fortran implement parallel algorithms by distributing large data structures across a multicomputer system. To enhance parallelism and reduce communication, it is sometimes beneficial for a programmer to change the distribution between phases of the algorithm. We introduce a new mapping strategy, called the spiral mapping, that reduces the communication overhead of array redistribution. Redistribution using the spiral mapping exploits communication locality and reduces global communication conflicts. We implemented redistribution using the standard linear mapping and the spiral mapping for two dimensional arrays;for 1024 x 1024 arrays, redistribution using the spiral mapping is 36% faster than using the linear mapping on a 16 node Intel iPSC/860.

关键词： HPF REDISTRIBUTION distributed memory computer PROCESSOR MAPPING

来源：评论

学校读者我要写书评

暂无评论

Parallel tools for colored image processing

引用

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE 1997年第7期11卷 1065-1084页

作者： Lozano, V Ubeda, S Vigouroux, X ECOLE NORMALE SUPER LYON LIP F-69364 LYON 7 FRANCE SITE GIAT IND F-42007 ST ETIENNE 1 FRANCE

Over recent years clusters of workstations have begun to replace systems of mainframes. These workstations can be used to behave as a parallel machine, and then could supply supercomputing performance at modest cost. We report an experiment of such a network within PVM framework for an industrial image analysis application. The goal of the application is to build an automated process of textile color pattern analysis. We plan to build several tools for parallel image analysis, and among those tools, we present here first the parallelization of a quantization algorithm and a way to parallelize a color image pyramid using the Parallel Virtual Machine environment. One important point is the use of the distributed storage that represents a set of workstations. This distribution allows the complete parallelization of the Read/Write operations.

关键词： distributed memory computer parallel file access color image processing multiresolution techniques

来源：评论

学校读者我要写书评

暂无评论

Parallel implementation of a large-scale 3-D air pollution model 3rd

引用

3rd International Conference on Large-Scale Scientific Computing (ICLSSC 2001)

作者： Ostromsky, T Zlatev, Z Bulgarian Acad Sci Cent Lab Parallel Processing BU-1113 Sofia Bulgaria Natl Environm Res Inst Dept Atmospher Environm DK-4000 Roskilde Denmark

ISBN: (纸本)3540430431

Air pollution models can efficiently be used in different environmental studies. The atmosphere is the most dynamic component of the environment, where the pollutants can be transported over very long distances. Therefore the models must be defined on a large space domain. Moreover, all relevant physical and chemical processes must be adequately described. This leads to huge computational tasks. That is why it is difficult to handle numerically such models even on the most powerful up-to-date supercomputers. The particular model used in this study is the Danish Eulerian Model. The numerical methods used in the advection-diffusion part of this model consist of finite elements (for discretizing the spatial derivatives) followed by predictor-corrector schemes with several different correctors (in the numerical treatment of the resulting systems of ordinary differential equations). Implicit methods for the solution of stiff systems of ordinary differential equations are used in the chemistry part. This implies the use of Newton-like iterative methods. A special sparse matrix technique is applied in order to increase the efficiency. The model is constantly updated with new faster and more accurate numerical methods. The three-dimensional version of the Danish Eulerian Model is presented in this work. The model is defined on a space domain of 4800 km x 4800 km that covers the whole of Europe together with parts of Asia, Africa and the Atlantic Ocean. A chemical scheme with 35 species is used in this version. Two parallel implementations are discussed;the first one for shared memory parallel computers, the second one - the newly developed version for distributed memory computers. Standard tools are used to achieve parallelism: OpenMP for shared memory computers and MPI for distributed memory computers. Results from many experiments, which were carried out on a SUN SMP cluster and on a CRAY T3E at the Edinburgh Parallel computer Centre (EPCC), are presented and analyzed.

关键词： air pollution model system of PDE's parallel algorithm shared memory computer distributed memory computer OpenMP MPI

来源：评论

学校读者我要写书评

暂无评论

Parallel implementation of minimum spanning tree algorithms using MPI

Parallel implementation of minimum spanning tree algorithms ...

引用

13th IEEE International Symposium on Computational Intelligence and Informatics (CINTI)

作者： Loncar, Vladimir Skrbic, Srdjan Univ Novi Sad Faculy Sci Depatment Math Informat Novi Sad 21000 Serbia

In this paper we study parallel algorithms for finding minimum spanning tree of a graph. We present two algorithms, based on sequential algorithms of Prim and Kruskal, targeting message passing parallel machine with d... 详细信息

ISBN: (纸本)9781467352062;9781467352055

关键词： Minimum spanning tree parallel algorithms message passing distributed memory computer

来源：评论

学校读者我要写书评

暂无评论

An automatic load-balanced parallel multilevel fast multipole method for large scale electromagnetic scattering problem

An automatic load-balanced parallel multilevel fast multipol...

引用

Asia-Pacific Microwave Conference

作者： Dong, J. Chai, S. L. Mao, J. J. Natl Univ Def Technol Sch Elect Sci & Engn Changsha 410073 HuNan Peoples R China

ISBN: (纸本)078039433X

A parallel multilevel fast multipole method for computing the large scale electromagnetic scattering problem is proposed and implemented in this paper. In recent years, multilevel fast multipole method (MLFMA) has been widely used for analyzing electromagnetic scattering problem of electric large object. In order to extend the range of problem that can be solved using this algorithm, we implement a parallel fast multipole method which can run on the distributed memory computer system. This parallel algorithm is based on the Message-passing Interface (MPI). A compressed octree based parallel domain subdivision algorithm is used for efficiently avoiding the load balance problem between different CPUs which is caused by the irregular object shape. The parallel efficiency of this algorithm is demonstrated by different computing examples. We have solved electromagnetic scattering problem of many practicality military targets with more than 5,000,000 unknowns using our parallel multilevel fast multipole code on Drawing 4000A Super-computer at Shanghai Super-Computing Center.

关键词： computational electromagnetics distributed shared memory systems electromagnetic wave scattering message passing military computing military radar octrees parallel algorithms resource allocation automatic multipole method distributed memory computer elec

来源：评论

学校读者我要写书评

暂无评论

SCHEDULING A SCATTERING-GATHERING SEQUENCE ON HYPERCUBES

引用

Parallel Processing Letters 1993年第1期3卷 29-42页

作者： CHARLES HENRI-PIERRE FRAIGNIAUD PIERRE Laboratoire de l’Informatique du Parallélisme—IMAG CNRS Ecole Normale Supérieure de Lyon 69364 Lyon Cedex 07 France Supported by the Basic Research Action ESPRIT 6632 “NANA2”. Supported by the Research Programs C3 and ANM and by the Direction des Recherches et Etudes Techniques.

The scattering problem refers to the gossiping and the broadcasting problems [1, 2]. It consists in distributing a set of data from a single source such that each component is sent to a distinct address. The gathering operation is the reverse of the scattering operation. This paper studies the problem of pipelining a scattering-gathering sequence in order to overlap these operations. We first give a general solution for distributed memory parallel computers, and next we particularly study this problem on hypercubes.

关键词： Broadcasting scattering gathering distributed memory computer

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：