检索结果-内蒙古大学图书馆

25th IEEE International Geoscience and Remote Sensing Symposium (IGARSS 2005)

作者： Fu, BT Wang, C Zhang, QW Huazhong Univ Sci & Technol Sch Hydropower & Informat Engn Wuhan 430074 Peoples R China

ISBN: (纸本)0780390504

Since MODIS data have the feature of huge capacity and multi-spectrum, it needs too much time and frequent I/O operations to correct them by RS software. This paper proposes a parallel algorithm for MODIS Data based on triangulation network. The input images are divided into several pieces and each CPU processes a piece independently. By implementing the algorithm on a cluster system, the results show that, this parallel algorithm improves the efficiency of geometrical correction greatly.

关键词： mODIS parallel algorithm geometrical correction triangulation network

来源：评论

学校读者我要写书评

暂无评论

parallel simulation of independent beam-tunnels in multiple-beam klystrons using TESLA

引用

IEEE TRANSACTIONS ON PLASMA SCIENCE 2008年第3期36卷 670-681页

作者： Chernyavskiy, Igor A. Cooke, Simon J. Vlasov, Alexander N. Antonsen, Thomas M., Jr. Abe, David K. Levush, Baruch Nguyen, Khanh T. Sci Applicat Int Corp Mclean VA 22102 USA USN Res Lab Washington DC 20375 USA Univ Maryland Inst Res Elect & Appl Phys College Pk MD 20742 USA Beam Wave Res Inc Bethesda MD 20814 USA

We present an extension of the klystron simulation code TESLA to model multiple-beam klystrons (MBKs) in which interaction parameters may vary significantly from beam-tunnel to beam-tunnel. In earlier work, the single-beam code was applied to model the MBK by assuming that all electron beams and beam-tunnels were identical and all electron beams interacted identically with the fields of the resonant cavities, using averaged values of R/Q to represent interaction with each resonant cavity. To overcome the limitations of this approach and to take into account the effects from nonidentical beams and/or beam-tunnels, we have modified the code to use a parallel algorithm for multiple beams. The implementation of the parallel version of TESLA is based on the latest Fortran-95 version of the serial code and uses the message-passing interface library for communication. For testing and verification purposes, the new version of the code is applied to simulate an experimental four-cavity, eight-beam klystron amplifier, which was designed and successfully tested last year at the Naval Research Laboratory. The results of modeling using the new parallel TESLA and their comparison with experimental data are discussed in detail.

关键词： large-signal code TESLA message-passing interface (MPI) multiple-beam klystron (MBK) parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

parallel numerical simulation with domain decomposition for the fluid-filled drum crash

引用

PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART C-JOURNAL OF MECHANICAL ENGINEERING SCIENCE 2008年第3期222卷 319-328页

作者： Li, Z. Jin, X-L Chen, X-D Shanghai Jiao Tong Univ Ctr High Performance Comp Shanghai 200240 Peoples R China

Fluid-structure interaction (FSI) problems simultaneously bring together some of the critical aspects associated with both fluid dynamics and structural dynamics. In this research, the simulation of the three-dimensional flexible fluid-filled drum in the crash is achieved through multi-material arbitrary Lagrangian-Eulerian (ALE) finite-element method because of its ability to control mesh geometry independently from geometry. The ALE description is adopted for the fluid domain, whereas for the structural domain the Lagrangian formulation is adopted. The computation of the FSI and the crash contact between the drum and the ground is realized by the penalty-based coupling method. Then the dynamic behaviour of the drum in the crash is analysed and the parallelism is discussed because the computation of the FSI and the crash contact is quite time-consuming. Based on domain decomposition, the recursive coordinate bisection (RCB) is improved according to the time-consuming characteristics of the fluid-filled container in the crash. The results indicate, in comparison with RCB method, the improved recursive coordinate bisection method has improved the speedup and the parallel efficiency.

关键词： fluid-filled container parallel algorithm arbitrary Lagrangian-Eulerian finite element domain decomposition recursive coordinate bisection

来源：评论

学校读者我要写书评

暂无评论

Fast electrostatic force calculation on parallel computer clusters

引用

JOURNAL OF COMPUTATIONAL PHYSICS 2008年第19期227卷 8551-8567页

作者： Kia, Amirali Kim, Daejoong Darve, Eric Stanford Univ Mech & Computat Div Stanford CA 94305 USA Sogang Univ Seoul 121742 South Korea

The fast multipole method (FMM) and smooth particle mesh Ewald (SPME) are well known fast algorithms to evaluate long range electrostatic interactions in molecular dynamics and other fields. FMM is a multi-scale method which reduces the computation cost by approximating the potential due to a group of particles at a large distance using few multipole functions. This algorithm scales like O(N) for N particles. SPME algorithm is an O(N In N) method which is based on an interpolation of the Fourier space part of the Ewald sum and evaluating the resulting convolutions using fast Fourier transform (FFT). Those algorithms suffer from relatively poor efficiency on large parallel machines especially for mid-size problems around hundreds of thousands of atoms. A variation of the FMM, called PWA, based on plane wave expansions is presented in this paper. A new parallelization strategy for PWA, which takes advantage of the specific form of this expansion, is described. Its parallel efficiency is compared with SPME through detail time measurements on two different computer clusters. (C) 2008 Elsevier Inc. All rights reserved.

关键词： fast multipole method particle mesh Ewald plane wave parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

A parallel adaptive finite-element package based on ALBERTA

引用

INTERNATIONAL JOURNAL OF COMPUTER MATHEMATICS 2008年第12期85卷 1793-1805页

作者： Liu, QingKai Mo, Zeyao Zhang, Linbo Chinese Acad Sci Acad Math & Syst Sci Beijing 100864 Peoples R China

ALBERTA, a sequential adaptive finite-element toolbox, is being used widely in the fields of scientific and engineering computation, especially in the numerical simulation of electromagnetics. But the nature of sequentiality has become the bottle-neck while solving large scale problems. So we develop a parallel adaptive finite-element package based on ALBERTA, using ParMETIS and PETSc. The package is able to deal with any problem that ALBERT solved. Furthermore, it is suitable for distributed memory parallel computers including PC clusters. In this paper, we present the implementation of the package in detail, and address several key algorithms and strategies of parallelization. Finally, some numerical experiments are given to show the performance and scalability of our package.

关键词： finite-element method adaption parallel algorithm ALBERTA

来源：评论

学校读者我要写书评

暂无评论

A NEW parallel SPATIAL QUERY algorithm FOR DISTRIBUTED SPATIAL DATABASES

A NEW PARALLEL SPATIAL QUERY ALGORITHM FOR DISTRIBUTED SPATI...

引用

2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)

作者： HONG-JUN WEI ZU-KUAN WEI QIAN YIN School of Computer Science and Engineering University of Electronic Science and Technology of China College of Information Science and Technology Beijing Normal University Beijing 100875 China

The balance of data and the utilization of resources are essential to distributed spatial database system. The paper presents an efficient parallel spatial query algorithm which takes seriously the organization of spatial data into account. The algorithm adopts a balanced spatial data partitioning strategy for distributed spatial databases. According to the characteristics of data partitioning, it builds a packing R-tree as its index. The strategy also considers the problem of computing distribution. By replicating index to every site, each site can access different entry in the same index node at the same time. Based on the organization of spatial data, the algorithm can simultaneously execute query operation at different site in both filtration phase and refinery phase. So it obviously improves spatial query performances. In order to solve multiple paths search problem caused by R-tree index, the algorithm brings in globe stack to buffer temporary index nodes. It settles the difficult problem flexibly in distributed spatial databases. For simplicity, the paper discusses the parallel algorithm in 2-dimensional space. Through the experiments conducting on many real datasets, it shows better performance in various spatial query operations.

关键词： Distributed spatial database Spatial query R-tree Globe stack parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

The Research and Realization of the parallel Gmres(m) algorithm

The Research and Realization of the Parallel Gmres(m) Algori...

引用

2008年国际电子商务、工程及科学领域的分布式计算和应用学术研讨会

作者： Aimin Yang~1 Shaohong Yan~1 Chunfeng Liu~(12,)Yamian Peng~1 Yanbin Sun~1 Yan Yan~1 1 College of Science,Hebei Polytechnic University,Tangshan,Hebei,063000,China 2 Hebei University of Technology,Tianjin,300130,China

With the rapid development of high-speed network technology,the cluster systems have been the main platform of parallel *** of the delay of their high communication,some parallel algorithms of fine grain are not fit to run in this environment. Therefore,it is necessary to study their parallel achievements in cluster *** terms of that,this paper aims at the internal parallel of the GMRES(m) method in order to find the solution of the linear equation groups and obtains coarse grain parallel algorithms,and more,we devise the program of this method using *** last,the example expresses that the designing parallel algorithm has much higher speedup in this cluster system.

关键词： Cluster GMRES(m) method parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

parallel randomized sampling for support vector machine (SVM) and support vector regression (SVR)

引用

KNOWLEDGE AND INFORMATION SYSTEMS 2008年第2期14卷 233-247页

作者： Lu, Yumao Roychowdhury, Vwani Yahoo Inc Sunnyvale CA 94089 USA Univ Calif Los Angeles Dept Elect Engn Los Angeles CA 90024 USA

A parallel randomized support vector machine (PRSVM) and a parallel randomized support vector regression (PRSVR) algorithm based on a randomized sampling technique are proposed in this paper. The proposed PRSVM and PRSVR have four major advantages over previous methods. (1) We prove that the proposed algorithms achieve an average convergence rate that is so far the fastest bounded convergence rate, among all SVM decomposition training algorithms to the best of our knowledge. The fast average convergence bound is achieved by a unique priority based sampling mechanism. (2) Unlike previous work (Provably fast training algorithm for support vector machines, 2001) the proposed algorithms work for general linear-nonseparable SVM and general non-linear SVR problems. This improvement is achieved by modeling new LP-type problems based on Karush-Kuhn-Tucker optimality conditions. (3) The proposed algorithms are the first parallel version of randomized sampling algorithms for SVM and SVR. Both the analytical convergence bound and the numerical results in a real application show that the proposed algorithm has good scalability. (4) We present demonstrations of the algorithms based on both synthetic data and data obtained from a real word application. Performance comparisons with SVMlight show that the proposed algorithms may be efficiently implemented.

关键词： randomized sampling support vector machine support vector regression parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

parallel randomized sampling for support vector machine (SVM) and support vector regression (SVR)

Parallel randomized sampling for support vector machine (SVM...

引用

10th Pacific-Asia Conference on Knowledge Discovery and Data Mining

作者： Lu, Yumao Roychowdhury, Vwani Yahoo Inc Sunnyvale CA 94089 USA Univ Calif Los Angeles Dept Elect Engn Los Angeles CA 90024 USA

关键词： randomized sampling support vector machine support vector regression parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

Design, Tuning and Evaluation of parallel Multilevel ILU Preconditioners

Design, Tuning and Evaluation of Parallel Multilevel ILU Pre...

引用

8th International Conference on High Performance Computing for Computational Science (VECPAR 2008)

作者： Aliaga, Jose I. Bollhoefer, Matthias Martin, Alberto F. Quintana-Orti, Enrique S. Univ Jaume 1 Depto Ingn & Ciencia Comp Castellon de La Plana 12071 Spain Tech Univ Carolo Wilhelmina Braunschweig Inst Computat Math D-38106 Braunschweig Germany

ISBN: (纸本)9783540928584

In this paper, we present a parallel multilevel TLU preconditioner implemented with OpenMP. We employ METIS partitioning algorithms to decompose the computation into concurrent tasks, which are then scheduled to threads. Concretely, we combine decompositions which obtain significantly more tasks than processors, and the use of dynamic scheduling strategies in order to reduce the thread's idle time, which it is shown to be the main source of overhead in our parallel algorithm. Experimental results on a shared-memory platform consisting of 1.6 processors report remarkable performance for our approach.

关键词： Sparse linear system incomplete LU factorization parallel algorithm OpenMP shared-memory multiprocessor

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：