检索结果-内蒙古大学图书馆

5th international conference on parallel processing, Euro-Par 1999

作者： Sagastizábal, Claudia A. Solodov, Mikhail V. INRIA-Rocquencourt BP 105 78153 Le Chesnay France UFRJ/COPPE-PESC CP 68501 Rio de Janeiro RJ 21945-970 Brazil Instituto de Matemática Pura e Aplicada Estrada Dona Castorina 110 Rio de Janeiro RJ 22460-320 Brazil

ISBN: (纸本)3540664432

In the parallel variable distribution (PVD) approach for solving optimization problems, the variables are distributed among parallel processors with each processor having the primary responsibility for updating its block of variables while allowing the remaining "secondary" variables to change in a restricted fashion along some easily computable directions. For constrained nonlinear programs, convergence in [4] was established in the special case of convex block-separable constraints. In [11], the PVD approach was extended to problems with general convex constraints by means of utilizing the projected gradient directions for the change of secondary variables. In this paper, we propose two new variants of PVD for the constrained case. For the case of block-separable constraints, we develop a parallel sequential quadratic programming algorithm. this is the first PVD-type method which does not assume convexity of the feasible set for convergence. For inseparable convex constraints, we propose a PVD method based on suitable approximate projected gradient directions. Using such approximate directions is especially important when the projection operation is computationally expensive. © Springer-Verlag Berlin Heidelberg 1999.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

parallel fuzzy learning 5th

Parallel fuzzy learning

引用

5th international Work-conference on Artificial and Natural Neural Networks (IWANN 99)

作者： Russo, M Univ Messina Dept Phys I-98166 Messina Italy Ist Nazl Fis Nucl Natl Inst Phys Sect Catania I-95129 Catania Italy

ISBN: (纸本)3540660690

GEFREX is a genetic-neuro-fuzzy tool able to extract very compact fuzzy rules. It is mainly based on the genetic paradigm. this paper shows the preliminary result obtained by the author regarding the parallelization of this tool. the hw platform consists of a cluster of 7 PC-based single pentium II 233 MHz workstations and 1 PC-based double pentium II 300 MHz processor server. the communication network is a fast-ethernet dedicated LAN. the results achieved are very promising. With up to seven processing elements, the speedup obtainable is linear.

关键词： Fuzzy inference

来源：评论

学校读者我要写书评

暂无评论

Modular Fortran 90 implementation of a parallel atmospheric general circulation model

Modular Fortran 90 implementation of a parallel atmospheric ...

引用

5th international Euro-Par conference

作者： Sawyer, W Takacs, L Molod, A Lucchesi, R NASA Goddard Space Flight Ctr Data Assimilat Off Greenbelt MD 20771 USA Univ Maryland Dept Meteorol College Pk MD 20742 USA

ISBN: (纸本)3540664432

In this paper we present software engineering aspects of a parallel atmospheric General Circulation Model (GCM) currently nearing completion at the NASA Data Assimilation Office. the code is scientifically closely related to the current Goddard Earth Observing System (GEOS 2.x) GCM. the new software design, however, is a radical departure from the previous Fortran 77 code, using Fortran 90 derived data types and modules to compartmentalize the implementation. In order to reduce turnaround time needed to facilitate scientific development, the GEOS 3.x implementation uses a message-passing paradigm. this allows the performance to scale consistently to a large number of processors on both an Origin 2000 and a Gray T3E. the Message-Passing Interface (MPI) is used for portability. the modularity and performance of the parallel GEOS 3.x code allow faster development of scientific improvements such as variable resolution grids and a new land surface model which are key to keeping the GCM competitive with other leading models.

关键词： Message passing

来源：评论

学校读者我要写书评

暂无评论

Skipper: A skeleton-based parallel programming environment for real-time image processing applications 5th

Skipper: A skeleton-based parallel programming environment f...

引用

5th international conference on parallel Computing Technologies, PaCT 1999

作者： Sérot, Jocelyn Ginhac, Dominique Dérutin, Jean-Pierre LASMEA UMR 6602-CNRS Campus des Cezeaux AubièreF-63177 France

ISBN: (纸本)3540663630

this paper presents SKiPPER, a programming environment dedicated to the fast prototyping of parallel vision algorithms on MIMD- DM platforms. SKiPPER is based upon the concept of algorithmic skele- tons, i.e. higher order program constructs encapsulating recurring forms of parallel computations and hiding their low-level implementation de- tails. Each skeleton is given an architecture-independent functional (but executable) specification and a portable implementation as a generic pro- cess template. the source program is a purely functional specification of the algorithm in which all parallelism is made explicit by means of com- posing instances of selected skeletons, each instance taking as parameters the application specific sequential functions written in C. SKiPPER compiles this specification down to a process graph in which nodes cor- respond to sequential functions and/or skeleton control processes and edges to communications. this graph is then mapped onto the target topology using a third-party CAD software (SynDEx). the result is a dead-lock free, optimized (but still portable) distributed executive, which SKiPPER finally turns into executable code for the target platform. the initial specification, written in ML language, can also be executed on any sequential platform to check the correctness of the parallel algorithm. the applicability of SKiPPER concepts and tools has been demonstrated by parallelising several realistic real-time vision applications both on a multi-DSP platform and a network of workstations. It is here illustrated with a real-time vehicle detection and tracking application. © Springer-Verlag Berlin Heidelberg 1999.

关键词： Specifications

来源：评论

学校读者我要写书评

暂无评论

A parallel branch-and-bound algorithm for the classification problem

引用

6th international conference on High Performance Computing, HiPC 1999

作者： Balev, Stefan Andonov, Rumen Freville, Arnaud Université de Valenciennes et du Hainaut-Cambresis LAMIH ROI UMR 8530 Le Mont HOUY 59304 - Valenciennes cedex France Sofia University Faculty of Mathematics and Computer Science Blvd. J. Bourchier 5 1164 Sofia Bulgaria

ISBN: (纸本)3540441271

the classification problem involves classifying an observation into one of two groups based on its attributes. Determination of a hyperplane which misclassifies the fewest number of observations from the training sample is a hard combinatorial optimization problem and the sequential algorithms for it are still not practical for large size instances. We propose a parallel branch-and-bound algorithm for the problem on distributed-memory MIMD architectures. It achieves a good load balance and a small communication overhead by using a simple load balancing scheme. Our approach is validated by experimental results. © 1999 Springer-Verlag Berlin Heidelberg.

关键词： Combinatorial optimization

来源：评论

学校读者我要写书评

暂无评论

SMARTS: Exploiting temporal locality and parallelism through vertical execution

Proceedings of the International Conference on Supercomputin...

引用

Proceedings of the international conference on Supercomputing 1999年 302-310页

作者： Vajracharya, Suvas Karmesin, Steve Beckman, Peter Crotinger, James Malony, Allen Shende, Sameer Oldehoeft, Rod Smith, Stephen Los Alamos Natl Lab Los Alamos United States

In the solution of large-scale numerical problems, parallel computing is becoming simultaneously more important and more difficult. the complex organization of today's multi-processors with several memory hierarchies has forced the scientific programmer to make a choice between simple but unscalable code and scalable but extremely complex code that does not port to other architectures. this paper describes how the SMARTS runtime system and the POOMA C++ class library for high-performance scientific computing work together to exploit data parallelism in scientific applications while hiding the details of managing parallelism and data locality from the user. We present innovative algorithms, based on the macro-dataflow model, for detecting data parallelism and efficiently executing data-parallel statements on shared-memory multiprocessors. We also describe how these algorithms can be implemented on clusters of SMPs.

关键词： parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Challenges for a real-world information processing by means of real-time neural computation and real-conditions simulation 5th

引用

5th international Work-conference on Artificial and Natural Neural Networks (IWANN 99)

作者： Herrero, JC Software AG Tres Cantos 28760 Madrid Spain

ISBN: (纸本)3540660682

Should we consider the dimensions of natural neural computation as they are known as a result of the scientific research, we realize there is a long tomorrow before us, interested in neural computation, for the simple reason that we can only handle a relatively low number of units and connections nowadays. All along this century we have significantly improved our knowledge on natural neural nets, to realize that huge number of cells and connections and begin to understand some of the brain signals processing and the repetitive structures which support it. However, even in the most developed cases, such as the auditory pathway modelling, there is not a neural computational device which can involve a real time response and follow the facts already known or plausibly postulated on some brain processes (e.g. by McCulloch and Pitts), with the unavoidable great number of processing elements involved too, besides neither suitable models regarding those kind of real-look nets have been designed nor their corresponding real-conditions simulations have been carried out. that means there is a lack of connectionistically computable models and also reduction methods by which we can obtain a connectionistic implementation design, given the knowledge level model. therefore, we would like to ask: what is within reach? In order to answer this question we are going to present a restricted auditory pathway modelling case, where we shall be able to see the realistic challenges we are facing up tp. By trying to propose a consistent implementation for it, based on parallel, modular, distributed and self-programming computation, we shall see the kind of methods, equipment, software and simulations required and desirable.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ultra-fast Poisson's equation solvers using wired-up processors for virtual devices architectures

Ultra-fast Poisson's equation solvers using wired-up process...

引用

international conference on Semiconductor

作者： S. Bara J. Zimmermann Lab. de Phys. des Composants a Semicond. CNRS Grenoble France

this paper presents two methods for solving a partial differential equation of the second order, with application to the well-known Poisson equation. these methods are aimed at making a high-speed hardware solver. the solutions presented will be a part of a hardware device simulator which is called "Virtual Device". We present simulation results to compare the two methods for solving this equation. We start with an iterative method (Gauss-Seidel method) and then end with a direct method (LU method).

关键词： Poisson equations Hardware parallel processing Iterative methods Gaussian processes Computational modeling Acceleration Iterative algorithms Partial differential equations Buildings

来源：评论

学校读者我要写书评

暂无评论

parallel fabrication of artificiallly designed superlattices by combinatorial laser MBE

引用

APPLIED PHYSICS A-MATERIALS SCIENCE & processing 1999年第1期69卷 S29-S31页

作者： Koinuma, H Koida, T Ohnishi, T Komiyama, D Lippmaa, M Kawasaki, M Tokyo Inst Technol Ceram Mat & Struct Lab Tokyo Japan Tokyo Inst Technol Interdisciplinary Grad Sch Sci & Engn Midori Ku Yokohama Kanagawa 2268503 Japan Japan Sci & Technol Corp CREST Yokohama Kanagawa Japan

A system for parallel synthesis of atomically regulated lattices and nano-structured heterojunctions has been developed. this combinatorial laser MBE system is composed of a UHV-PLD chamber, scanning RHEED for in situ diagnostics of reactions at various areas of substrate surface, a pair of fixed and moving masks to define the film deposition area, and a computer for controlling the operation conditions as functions of deposition area and time. In this system, oxide layered lattice strips were deposited in parallel on a SrTiO3 substrate under RHEED observation of individual strips with different compositions and/or layer thickness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

parallel architectures for image processing

引用

ELECTRONICS & COMMUNICATION ENGINEERING JOURNAL 1998年第3期10卷 139-151页

作者： Downton, A Crookes, D Univ Essex Dept Elect Syst Engn Colchester CO4 3SQ Essex England Queens Univ Belfast Dept Comp Sci Belfast BT7 1NN Antrim North Ireland

Image processing is often considered a good candidate for the application of parallel processing because of the large volumes of data and the complex algorithms commonly encountered. this paper presents a tutorial introduction to the field of parallel image processing. After introducing the classes of parallel processing a brief review of architectures for parallel image processing is presented. Software design for low-level image processing and parallelism in high-level image processing are discussed and an application of parallel processing to handwritten postcode recognition is described. the paper concludes with a look at future technology and market trends.

关键词： image processing image recognition parallel architecture Digital signal processing chips software design Microprocessors and microcomputers parallel architectures handwritten postcode recognition Optical information, image and video signal processing digital signal processing chips optical character recognition Computer vision and image processing techniques high-level image processing low-level image processing parallel processing

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：