检索结果-内蒙古大学图书馆

An assessment of some preconditioning techniques in shell problems

International Journal for Numerical Methods in Biomedical Engineering 1999年第10期14卷

作者： Michele Benzi Reijo Kouhia Miroslav Tůma Scientific Computing Group CIC-19 Los Alamos National Laboratory Los Alamos NM 87545 U.S.A. Laboratory of Structural Mechanics Helsinki University of Technology P.O. Box 2100 02015 HUT Finland Institute of Computer Science Czech Academy of Sciences 182 07 Prague 8 Czech Republic

Preconditioned Krylov subspace methods have proved to be efficient in solving large, sparse linear systems in many areas of scientific computing. The success of these methods in many cases is due to the existence of good preconditioning techniques. In problems of structural mechanics, like the analysis of heat transfer and deformation of solid bodies, iterative solution of the linear equation system can result in a significant reduction of computing time. Also many preconditioning techniques can be applied to these problems, thus facilitating the choice of an optimal preconditioning on the particular computer architecture available. However, in the analysis of thin shells the situation is not so transparent. It is well known that the stiffness matrices generated by the FE discretization of thin shells are very ill-conditioned. Thus, many preconditioning techniques fail to converge or they converge too slowly to be competitive with direct solvers. In this study, the performance of some general preconditioning techniques on shell problems is examined. © 1998 John Wiley & Sons, Ltd.

关键词： preconditioning conjugate gradient finite elements shells

来源：评论

学校读者我要写书评

暂无评论

PRCS: The project revision control system 8th

PRCS: The project revision control system

引用

8th Symposium on System Configuration Management, SCM 1998

作者： Macdonald, Josh Hilfinger, Paul N. Semenzato, Luigi University of California at Berkeley Department of Electrical Engineering and Computer Sciences BerkeleyCA94720 United States National Energy Research Scientific Computing Center Lawrence Berkeley National Laboratory BerkeleyCA94720 United States

ISBN: (纸本)3540647333

PRCS is an attempt to provide a version-control system for collections of files with a simple operational model, a clean user interface, and high performance. PRCS is characterized by the use of project description files to input most commands, instead of a point-and-click or a line-oriented interface. It employs optimistic concurrency control and encourages operations on the entire project rather than individual files. Although its current implementation uses RCS in the back-end, the interface completely hides its presence. PRCS is free. This paper describes the advantages and disadvantages of our approach, and discusses implementation issues. © Springer-Verlag Berlin Heidelberg 1998.

关键词： User interfaces

来源：评论

学校读者我要写书评

暂无评论

Visualization of scalar topology for structural enhancement

Visualization of scalar topology for structural enhancement

引用

IEEE Conference on Visualization

作者： C.L. Bajaj V. Pascucci D.R. Schikore Department of Computer Sciences and TICAM University of Technology Austin TX USA Lawrence Livermore National Laboratory Center for Applied Scientific Computing Livermore CA USA

Scalar fields arise in every scientific application. Existing scalar visualization techniques require that the user infers the global scalar structure from what is frequently an insufficient display of information. We present a visualization technique which numerically detects the structure at all scales, removing from the user the responsibility of extracting information implicit in the data, and presenting the structure explicitly for analysis. We further demonstrate how scalar topology detection proves useful for correct visualization and image processing applications such as image co-registration, isocontouring, and mesh compression.

关键词： Topology Data visualization Displays Data mining scientific computing Laboratories Application software Information analysis Image processing Medical services

来源：评论

学校读者我要写书评

暂无评论

Hypervolume visualization: a challenge in simplicity

Hypervolume visualization: a challenge in simplicity

引用

IEEE Symposium on Volume Visualization and Graphics

作者： C.L. Bajaj V. Pascucci G. Rabbiolo D.R. Schikore Department of Computer Sciences and TICAM University of Technology Austin TX USA Department of Mathematics Purdue University West Lafayette IN USA Center for Applied Scientific Computing Lawrence Livemore National Laboratory Livermore CA USA

Hypervolume visualization is designed to provide simple and fully explanatory images that give comprehensive in-sights into the global structure of scalar fields of any dimension. The basic idea is to have a dimension independent viewing system that scales nicely with the geometric dimension of the dataset and that can be combined with classical approaches like isocontouring and animation of slices of nD data. One completely abandons (for core simplicity) rendering techniques, such as hidden surface removal or lighting or radiosity, that enhance three dimensional realism and concentrate on the real-time display of images that highlight structural (topological) features of the no dataset (holes, tunnels, cavities, depressions, extrema, etc.). Hypervolume visualization on the one hand is a generalization of direct parallel projection methods in volume rendering. To achieve efficiency (and real-time performance on a graphics workstation) the authors combine the advantages of (i) a hierarchical representations of the hypervolume data for multiresolution display and (ii) generalized object space splatting combined with texture-mapped graphics hardware acceleration. The main results of the paper are thus both a multiresolution direct rendering algorithm and scalable graphical user interface that provides global views of scalar fields in any dimension, while maintaining the fundamental characteristics of ease of use, and quick exploratory user interaction.

关键词： Rendering (computer graphics) Graphics Animation Three dimensional displays Data visualization Workstations Hardware Acceleration Graphical user interfaces

来源：评论

学校读者我要写书评

暂无评论

Adaptive blocks: A high performance data structure 97

Adaptive blocks: A high performance data structure

引用

1997 ACM/IEEE Conference on Supercomputing, SC 1997

作者： Stout, Quentin F. De Zeeuw, Darren L. Gombosi, Tamas I. Groth, Clinton P. T. Marshall, Hal G. Powell, Kenneth G. Electrical Engineering and Computer Science Center for Parallel Computing University of Michigan Ann Arbor MI 48109-2122 United States Space Physics Research Laboratory Department of Atmospheric Oceanic and Space Sciences University of Michigan Ann Arbor MI 48109-2143 United States Laboratory for Scientific Computation Center for Parallel Computing University of Michigan Ann Arbor MI 48109-2094 United States Department of Aerospace Engineering University of Michigan Ann Arbor MI 48109-2118 United States

ISBN: (纸本)0897919858

We examine a data structure which uses flexible "adaptivity" to obtain high performance for both serial and parallel computers. The data structure is an adaptive grid which partitions a given region into regular cells. Its closest relatives are cell-based tree decompositions, but there are several important differences which lead to significant performance advantages. Using this block data structure to support adaptive mesh refinement (AMR), we were able to sustain 17 GFLOPS in ideal magnetohydrodynamic (MHD) simulations of the solar wind emanating from the base of the solar corona, using a 512 processor Cray T3D at NASA Goddard. © 1997 ACM.

关键词： Magnetohydrodynamics

来源：评论

学校读者我要写书评

暂无评论

Job management requirements for NAS parallel systems and clusters

引用

Workshop on Job Scheduling Strategies for Parallel Processing held as a prelude to the IPPS 1995 Conference

作者： Saphir, William Tanner, Leigh Ann Traversat, Bernard Computer Sciences Corporation NASA Contract NAS 2-12961 Moffett FieldCA94035-1000 United States NAS Scientific Computing Branch NASA Ames Research Center Mail Stop 258-6 Moffett FieldCA94035-1000 United States

ISBN: (纸本)3540601538

A job management system is a critical component of a production supercomputing environment, permitting oversubscribed resources to be shared fairly and efficiently. Job management systems that were originally designed for traditional vector supercomputers are not appropriate for the distributed-memory parallel supercomputers that are becoming increasingly important in the high performance computing industry. Newer job management systems offer new functionality but do not solve fundamental problems. We address some of the main issues in resource allocation and job scheduling we have encountered on two parallel computers - a 160-node IBM SP2 and a cluster of 20 high performance workstations located at the Numerical Aerodynamic Simulation facility. We describe the requirements for resource allocation and job management that are necessary to provide a production supercomputing environment on these machines, prioritizing according to difficulty and importance, and advocating a return to fundamental issues. © 1995, Springer Verlag. All rights reserved.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Reply to 'comment on 'poisson schemes for hamiltonian systems on poisson manifolds''

引用

computers & Mathematics with Applications 1995年第7期29卷 1-1页

作者： Zhu, W.-J. Qin, M.-Z. Department of Computer Science University of Minnesota Minneapolis MN 55455-0159 U.S.A. State Key Laboratory of Scientific and Engineering Computing Institute of Computational Mathematics Chinese Academy of Sciences Beijing 100080 P.R. China

来源：评论

学校读者我要写书评

暂无评论

Parallel solutions to the phase problem in X-ray crystallography: An update 6

Parallel solutions to the phase problem in X-ray crystallogr...

引用

6th Distributed Memory computing Conference, DMCC 1991

作者： De Titta, G. Sabin, T. Hauptman, H. Thuman, P. Miller, R. Weeks, C. Pagels, M. Graduate Group in Advanced Scientific Computing 226 Bell Hall University at Buffalo BuffaloNY14260 United States Medical Foundation of Buffalo 73 High Street BuffaloNY14203 United States Department of Chemistry University at Buffalo BuffaloNY14214 United States School of Medicine and Biophysical Sciences University at Buffalo BuffaloNY14214 United States Department of Computer Science University at Buffalo BuffaloNY14260 United States Department of Computer Science University of Arizona TucsonAZ85721 United States

ISBN: (纸本)0818622903

The central problem of single crystal molecular structure determination via x-ray diffraction is the "phase problem."Associated with each diffraction maximum are a magnitude, which can be experimentally determined, and a phase angle, which is lost in the experiment. The goal of "direct methods" is to mathematically reconstruct the phase information from the magnitude information alone. The traditional approach of direct methods is capable of determining structures of moderate complexity. The extension of this approach to problems of the size of macromolecules is hampered by the inherent limitations of the traditional approach. Recently, a new formulation of the "phase problem," dependent on a minimal function, has been proposed. This requires the determination of a set of phases which minimize the function. This work is a continuation of [1]. In this paper, we present the Shake-and-Bake method of structure determination. This is a new process aimed at determining structures via minimization and repeated Fourier transformations between reciprocal and real space. Implementation details on a network of Sun4 workstations will be given. In addition results will given for structures of sizes 28 and 84 atoms. © 1991 IEEE.

关键词： X ray crystallography

来源：评论

学校读者我要写书评

暂无评论

Parallel solutions to the phase problem in X-ray crystallography 5

Parallel solutions to the phase problem in X-ray crystallogr...

引用

5th Distributed Memory computing Conference, DMCC 1990

作者： Bashir, N. Hauptman, H. Miller, R. Crovella, M. Horvath, J. Sabin, T. De Titta, G. King, H. Thuman, P. Han, F. Langs, D. Velmurugan, D. Department of Computer Science University at Buffalo BuffaloNY14260 United States Department of Computer Science University at Buffalo United States Medical Foundation of Buffalo 73 High Street BuffaloMY14203 United States Department of Chemistry University at Buffalo BuffaloNY14214 United States Upjohn Company KalamazooMI49001 United States School of Medicine and Biophysical Sciences University at Buffalo BuffaloNY14214 United States Department of Computer Science University of Rochester RochesterNY14627 United States Department of Computer Science University of Wisconsin MadisonWI53706 United States Graduate Group in Advanced Scientific Computing University at Buffalo 226 Bell Hall BuffaloNY14260 United States

ISBN: (纸本)0818621133

The central problem of single crystal molecular structure determination via X-ray diffraction is the "phase problem." Associated with each diffraction maximum (a reflection) are a magnitude, which can be experimentally determined, and a phase angle, which is lost in the experiment. The goal of "direct methods" is to mathematically reconstruct the phase information from the magnitude information alone. Traditional direct methods are capable of determining structures of moderate complexity, but to extend them to problems of the size of macromolecules (proteins, etc.) requires developing new techniques that appear to be computationally intensive. Recently, a new formulation of the phasing process, dependent on a minimal function, has been proposed. Here we explore a number of different implementations of the principle to the solution of small molecular structures. The machines that we use include an Intel iPSC/2 hypercube, the Connection Machine CM- 2, and a network of Sun workstations. © 1990 IEEE.

关键词： Molecular structure

来源：评论

学校读者我要写书评

暂无评论

Communication Lower Bounds and Optimal Algorithms for Symmetric Matrix Computations

引用

ACM Transactions on Parallel computing 1000年

作者： Hussam Al Daas Grey Ballard Laura Grigori Suraj Kumar Kathryn Rouse Mathieu Verite STFC Scientific Computing Department Rutherford Appleton Laboratory Didcot United Kingdom of Great Britain and Northern Ireland Computer Science Wake Forest University Winston-Salem United States Institute of Mathematics EPFL Lausanne Switzerland PSI Center for Scientific Computing Theory and Data Villigen Switzerland Institut national de recherche en sciences et technologies du numérique Lyon France Inmar Intelligence Winston-Salem United States

In this article, we focus on the communication costs of three symmetric matrix computations: i) multiplying a matrix with its transpose, known as a symmetric rank-k update (SYRK) ii) adding the result of the multiplication of a matrix with the transpose of another matrix and the transpose of that result, known as a symmetric rank-2k update (SYR2K) iii) performing matrix multiplication with a symmetric input matrix (SYMM). All three computations appear in the Level 3 Basic Linear Algebra Subroutines (BLAS) and have wide use in applications involving symmetric matrices. We establish communication lower bounds for these kernels using sequential and distributed-memory parallel computational models, and we show that our bounds are tight by presenting communication-optimal algorithms for each setting. Our lower bound proofs rely on applying a geometric inequality for symmetric computations and analytically solving constrained nonlinear optimization problems. The symmetric matrix and its corresponding computations are accessed and performed according to a triangular block partitioning scheme in the optimal algorithms.

关键词：

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：