检索结果-内蒙古大学图书馆

Self-adapting numerical software (SANS) effort

IBM JOURNAL OF RESEARCH AND DEVELOPMENT 2006年第2-3期50卷 223-238页

作者： Dongarra, J Bosilca, G Chen, Z Eijkhout, V Fagg, GE Fuentes, E Langou, J Luszczek, P Pjesivac-Grbovic, J Seymour, K You, H Vadhiyar, SS Univ Tennessee Dept Comp Sci Innovat Comp Lab Knoxville TN 37996 USA Indian Inst Sci Supercomp Educ & Res Ctr Bangalore 560012 Karnataka India

The challenge for the development of next-generation software is the successful management of the complex computational environment while delivering to the scientist the full power of flexible compositions of the available algorithmic alternatives. Self-adapting numerical software (SANS) systems are intended to meet this significant challenge. The process of arriving at an efficient numerical solution of problems in computational science involves numerous decisions by a numerical expert. Attempts to automate such decisions distinguish three levels: algorithmic decision, management of the parallel environment, and processor-specific tuning. of kernels. Additionally, at any of these levels we can decide to rearrange the users data. In this paper we look at a number of efforts at the University of Tennessee to investigate these areas.

关键词： .orts user data Self-adapting linear systems numerical software Innovative Computing

来源：评论

学校读者我要写书评

暂无评论

Improving numerical software

引用

APPLIED MATHEMATICS AND COMPUTATION 2007年第1期192卷 247-251页

作者： Zotos, Kostas Univ Macedonia Dept Appl Informat Thessaloniki Greece

Traditionally there have been two opposite theories in software engineering. The first is that users want "black box'' software that they can use with complete confidence for general problem classes without having to understand the. ne algorithmic details. The second is that users want to be able to tune data structures for a particular application, even if the software is not as reliable as that provided for general methods. It turns out both are true, for different groups of users. Traditionally, users have asked for and been provided with "black box'' software in the form of mathematical libraries such as LAPACK, LINPACK, NAG, and IMSL. More recently, the high-performance community has discovered that they must write custom software for their problem. Their reasons include inadequate functionality of existing software libraries, data structures that are not natural or convenient for a particular problem and overly general software that sacrifices too much performance when applied to a special case of interest. Can we improve numerical libraries? The answer is yes and we are going to investigate in this paper some methods. (C) 2007 Elsevier Inc. All rights reserved.

关键词： numerical software performance of OO software improving OO software

来源：评论

学校读者我要写书评

暂无评论

An open question to developers of numerical software

引用

COMPUTER 2005年第5期38卷 91-94页

作者： Kahan, W Zuras, D Univ Calif Berkeley Berkeley CA 94720 USA

IEEE 754 a standard for binary floating-point arithmetic has revolutionized the portability and reliability of programs that use binary floating-point arithmetic. Floating point is almost universally implemented with special-purpose hardware that tucks into a small corner of the CPU chip and runs in the hundreds of Mflops to Gflops range. Single-stepping through today's floating-point software to debug it often turns out to be futile. The concept of a NaN, standing for "not a number", evolved from an "indefinite" in Seymour Cray's CDC 6600. IEEE 754, by default, requires an untrapped "invalid operation", to signal itself by raising a flag and to deliver a NaN just when any other result, be it finite or infinite, would cause worse confusion. The NaN lets a program retain control unless the program or programmer directs its cancellation upon an invalid operation. Thus, a program conducting a search can return to the realm being searched after an accidental foray beyond a boundary whose existence and location were previously unknown. A sNaN differs from the other quiet NaNs by traooing any attempt to perform arithmetic upon it; then a trap-handler must interpret this sNaN.

关键词： CPU chip IEEE 754 IEEE standards binary floating-point arithmetic standard computing milieux error analysis floating point arithmetic mathematics computing numerical software numerical software development program control sNaN sNaNs special-purpose hardware standards

来源：评论

学校读者我要写书评

暂无评论

UNDERFLOW AND THE RELIABILITY OF numerical software

引用

SIAM JOURNAL ON SCIENTIFIC AND STATISTICAL COMPUTING 1984年第4期5卷 887-919页

作者： DEMMEL, J

We examine the effects of different underflow mechanisms on the reliability of numerical software. software is considered reliable in the face of underflow if the effects of underflow are no worse than the uncertainty due to roundoff alone. The two primary underflow mechanisms discussed are store zero and gradual underflow, although we consider other mechanisms as well. By examining a variety of codes, including Gaussian elimination, polynomial evaluation, and eigenvalue calculation, we conclude that gradual underflow makes it significantly easier to write good numerical codes than store zero, and that this remains true even if extra range and precision are available for intermediate calculations.

关键词： software reliability numerical software roundoff underflow error analysis

来源：评论

学校读者我要写书评

暂无评论

Spin algebra and NMR theory using numerical software

引用

CONCEPTS IN MAGNETIC RESONANCE PART A 2015年第2期44卷 114-132页

作者： Grivet, Jean-Philippe Univ Orleans Dept Phys Orleans France CNRS Ctr Biophys Mol Orleans France

A simple method to execute spin algebra computations is presented. It uses the high-level numerical program Scilab and its large and convenient set of linear algebra routines. The techniques thus developed are applied to various nuclear magnetic resonance (NMR)-related simulations. Static spectra of AB and ABC spin systems are computed. Free precession of single or multispin systems are simulated using the density matrix formalism. Magnetization transfer under simple pulse sequences is examined. Matrices representing product operators are displayed and coherence orders are explained. COSY correlation maps for two and three spins are obtained without approximation. The simulation of a DQF-COSY spectrum, with its eight-step phase program, is explained. Two simple pulsed field gradient experiments are described. Computation times are under 100 s on a 1.66-GHz laptop computer. The approach is elementary, requiring only a basic knowledge of spin quantum mechanics. (c) 2015 Wiley Periodicals, Inc. Concepts Magn Reson Part A 44A: 114-132, 2015.

关键词： spin algebra numerical software Scilab spectral simulation density matrix high resolution NMR spectra free precession magnetization transfer product operators coherence orders COSY maps DQF-COSY pulsed field gradients

来源：评论

学校读者我要写书评

暂无评论

3D Modeling of Complicated Geological Body in numerical software

3D Modeling of Complicated Geological Body in Numerical Soft...

引用

International Conference on Civil Engineering and Building Materials (CEBM)

作者： Yu, Dongming Yao, Hailin Chen, Fengguang Chinese Acad Sci Inst Rock & Soil Mech Wuhan 430071 Peoples R China

ISBN: (纸本)9783037851395

This paper provides a simple and effective modeling method for complicated geological bodies in numerical software. By means of the most common tool software, the high quality 3d numerical model of the complicated geological body in the geotechnical engineering was built quickly and easily in the numerical software, ANSYS. Furthermore, this numerical model was reproduced in other numerical software, FLAC(3D) and ABAQUS. It is achieved that the 3d numerical model is shared among the numerical software, ANSYS, FLAC(3D) and ABAQUS and a basis is supplied for numerical calculation. The modeling method in this paper is of practical significance.

关键词： numerical simulation Model Complicated geological body numerical software

来源：评论

学校读者我要写书评

暂无评论

The fundamental solution for shallow circular cylindrical shells - Part II: numerical computation and software

引用

INTERNATIONAL JOURNAL OF ENGINEERING SCIENCE 2000年第11期38卷 1259-1274页

作者： Chen, G Coleman, MP Morris, PJ Texas A&M Univ Dept Math College Stn TX 77843 USA Fairfield Univ Dept Math & Comp Sci Fairfield CT 06430 USA Penn State Univ Dept Aerosp Engn University Pk PA 16802 USA

The explicit closed form fundamental solution of the elastostatic shallow circular cylindrical shell derived in Part I by Chen et al. [Int. J. Eng, Sci, to appear] is realized numerically. Major difficulties exist in approximating integrals of Hankel functions with irrational functions as arguments on the two-dimensional domain of definition. By analyzing the asymptotic characteristics of those integrals, we decompose the first quadrant of the (x, s)-plane into five regions, and apply appropriate exact, series-truncation, or quadrature algorithms on each region to compute numerical values of the three components of the fundamental solution. Comparisons of errors are tabulated and graphics of solutions are illustrated. (C) 2000 Elsevier Science Ltd. All rights reserved.

关键词： shallow circular cylindrical shell fundamental solution numerical software

来源：评论

学校读者我要写书评

暂无评论

Object-Oriented numerical software Trilinos and Its Application

Object-Oriented Numerical Software Trilinos and Its Applicat...

引用

2006年国际电子、工程及科学领域的分布式计算和应用学术研讨会

作者： Weiting An~(1,2),Jianwen Cao~2,Ni Zhang~(1,2) 1 The Graduate University of Chinese Academy of Sciences 2 Laboratory of Parallel Computing,Institute of software,CAS Beijing,100080,China

Trilinos is Object-Oriented numerical *** is an effort to facilitate the design,development,integration,and ongoing support of mathematical software *** this paper we firstly present an introduction to Trilinos,its linear package Epetra and solver package AztecOO,and then we use Trilinos to solve linear systems from Matrix Market database on a parallel *** sparse matrices are partitioned using *** linear solver we used is from *** hMETIS-Epetra-AztecOO method combines hMETIS,Epetra,and AztecOO together and can be used in many other scientific and engineering applications.

关键词： numerical software parallel computing iterative methods Object-Oriented technology data partitioning

来源：评论

学校读者我要写书评

暂无评论

numerical Algorithms and software Tools for Analysis and Modelling of Descriptor Systems

引用

IFAC Proceedings Volumes 1992年第21期25卷 392-395页

作者： Andras Varga Department of Mechanical Engineering Ruhr-University Bochum Institute for Automatic Control P.O. Box 102148 D-4630 Bochum Germany

The paper presents a state-of-the art overview of algorithms and software for analysis and modelling of linear descriptor systems. The presentation focuses on algorithms for which reliable software implementations are already available. Most of mentioned software tools belong to a small Fortran subroutines library, called DESCRIPT, developed by the author. This library contains robust implementation of numerically reliable algorithms covering the basic computational problems in analysis and mdelling of descriptor systems. DESCRIPT is the first systematized collection of software tools dedicated, to descriptor systems. The analysis and modelling facilities for descriptor systems available in the interactive package CASAM are based exclusively on the subroutines from DESCRIPT.

关键词： pescriptor systems numerical methods Systems analysis numerical software

来源：评论

学校读者我要写书评

暂无评论

Symbolic testing of floating-point bugs and exceptions

引用

JOURNAL OF SYSTEMS AND software 2025年 219卷

作者： Ma, Dongyu Liang, Zeyu Yin, Luming Liang, Hongliang Beijing Univ Posts & Telecommun Beijing Peoples R China Beijing Univ Posts & Telecommun Trusted Software & Intelligent Syst Res Grp Beijing Peoples R China Univ Calif San Diego La Jolla CA USA

numerical software are susceptible to floating-point bugs and exceptions, which may lead to severe threats like denial of service attacks. Static analysis techniques such as symbolic execution are effective in detecting general bugs which often cause memory error or program crash. Unfortunately, these methods do not deal well with numerical code as they do not support floating-point constraints and math functions symbolically. In this paper, we propose a new analysis framework YUSE, which can detect floating-point bugs by constructing constraints and exploring paths which contain floating-point expressions. Specifically, we introduce interval computation and interval constraint propagation in non-relational numerical abstract domains, and symbolically model math functions, to accurately detect floating-point bugs and exceptions. Moreover, we leverage two-phase constraint solving to enhance YUSE's performance. Experimental results show that YUSE outperforms two state-of-the-art tools, Frama-c and Fpse-study, in terms of effectiveness and efficiency, with 1.4x x and 7.1x x faster than Frama-c and Fpse-study, respectively. Moreover, YUSE found 20 new bugs in real-world software, 12 of which were assigned CVE IDs and 8 of which were confirmed by developers.

关键词： numerical software Floating-point exceptions and bugs Symbolic execution

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：