检索结果-内蒙古大学图书馆

International Symposium on parallel and Distributed Processing (IPDPS)

作者： E. Alba G. Luque Departamento de Lenguajes y Ciencias de la Computación E.T.S. Ingeniería Informática Malaga Spain

We present in this work a wide spectrum of results on analyzing the behavior of parallel heuristics for solving optimization problems. We focus on evolutionary algorithms as well as on simulated annealing. Our goal is to offer a first study on the possible changes in the search mechanics when shifting from a LAN distributed algorithm to a WAN environment. We address six optimization tasks of considerable complexity. The results show that, despite the expected slower execution time, the WAN versions of our algorithms consistently solve the problems. We even report some interesting results in which WAN algorithms outperform LAN ones. We also extend the study to include hybrid versions to check the scope of our conclusions.

关键词： Local area networks Wide area networks Skeleton Algorithm design and analysis parallel algorithms Evolutionary computation Computational modeling Simulated annealing Distributed algorithms Design optimization

来源：评论

学校读者我要写书评

暂无评论

A new data distribution method for parallel ray tracing

A new data distribution method for parallel ray tracing

引用

International Conference on Information Visualisation (IV)

作者： M. Amri BP-SONARCO-Information System Department-Hassi Messaoud Houari Boumedienne-Algiers University of Science and Technology Algeria

The ray tracing algorithm produces realistic high quality images, but it requires a long time for calculation on a single processor machine, which limits its practical use. With the development of distributed object architectures, such as CORBA (Common Object Request Broker), the most promising way to improve ray traced picture productions seems to be parallelisation, which offers both increased CPU power and memory facilities. A natural way for parallelisation is to distribute pixels over the CORBA object system. However, since we want to deal with large scenes and making our algorithm functional over any parallel architecture system, scene objects have also to be distributed among processors, so a modified parallel algorithm is necessary. We propose a new method, which distributes the scene objects among processors (CORBA objects) according to their speed frequencies, which uses the object coherency property. Our approach of exploiting the bus CORBA gives very encouraging results.

关键词： Ray tracing Layout Network servers Computer graphics Computer architecture Central Processing Unit Information systems Production parallel architectures parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

Proceedings International parallel and Distributed Processing Symposium

Proceedings International Parallel and Distributed Processin...

引用

International Symposium on parallel and Distributed Processing (IPDPS)

The following topics are dealt with: Grid and distributed computing; scheduling task systems; shared-memory multiprocessors; imaging and visualization; testing and debugging; performance analysis and real-time systems; scheduling for heterogeneous resources; networking; peer-to-peer and mobile computing; compiler technology and run-time systems; load balancing; network routing; parallel programming models; parallel algorithms; scheduling and storage; parallel and distributed performance; software for high performance clusters; decentralized algorithms; multithreading and VLIW; parallel and distributed real-time systems; high-level parallel programming models and supportive environments; Java for parallel and distributed computing; nature inspired distributed computing; high performance computational biology; advances in parallel and distributed computational models; reconfigurable architectures; communication architecture for clusters; next generation systems; fault-tolerant parallel and distributed systems; wireless, mobile and ad hoc networks; parallel and distributed image processing, video processing, and multimedia; formal methods for parallel programming; Internet computing and e-commerce; parallel and distributed scientific and engineering computing with applications; massively parallel processing; performance modeling, evaluation, and optimization of parallel and distributed systems; and parallel and distributed systems: testing and debugging.

关键词： Processor scheduling Shared memory systems Software debugging Software testing Real time systems Program compilers Resource management parallel programming parallel algorithms Memory management parallel architectures Biomedical computing Reconfigurable architectures Computer fault tolerance Video signal processing Image processing Software requirements and specifications Multimedia computing Internet

来源：评论

学校读者我要写书评

暂无评论

A parallel algorithm for computing simultaneous inversions with application to elliptic curve scalar multiplication

A parallel algorithm for computing simultaneous inversions w...

引用

Midwest Symposium on Circuits and Systems (MWSCAS)

作者： P. Sarkar P.K. Mishra R. Barua Cryptology Researchi Group Applied Statistics Unit Indian Statistical Institute Calcutta India

Montgomery's trick is a well known technique for performing simultaneous inversions of several field elements. However, this technique is a strictly sequential algorithm. Here the authors introduced a parallel algorithm for performing simultaneous inversions of several finite field elements. The algorithm uses a binary tree and can perform inversions of 2 r elements using 3times2 r-1 multipliers in (r + 1) multiplication rounds and one inversion round. The authors also described how to modify the algorithm when less number of multipliers is available. This parallel algorithm is used to obtain a new parallel algorithm for elliptic curve scalar multiplication using a fixed base point. The scalar multiplication algorithm is resistant against simple power analysis (SPA) and can be implemented with different number of multipliers (2,4,8,...). Results show that implementation with 2 multipliers can lead to almost 40% speed-up over previously best known sequential SPA resistant algorithm

关键词： parallel algorithms Concurrent computing Elliptic curves Galois fields Elliptic curve cryptography Costs Binary trees Algorithm design and analysis Statistics Indium phosphide

来源：评论

学校读者我要写书评

暂无评论

Optimal parallel algorithm for the knapsack problem without memory conflicts

Optimal parallel algorithm for the knapsack problem without ...

引用

IEEE International Conference on parallel and Distributed Computing, Applications and Technologies (PDCAT)

作者： Kenli Li Qinghua Li Wang-Hui Shengyi Jiang School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China National High Performance Computing Center Wuhan China

The knapsack problem is very important in cryptosystem and in number theory. We propose a new parallel algorithm for the knapsack problem where the method of divide and conquer is adopted. Basing on an EREW-SIMD machine with shared memory, the proposed algorithm utilizes O(2/sup n/4/)/sup 1-/spl epsiv// processors, 0/spl les//spl epsiv//spl les/1, and O(2/sup n/) memory to find a solution for the n-element knapsack problem in time O(2/sup n/4/ (2/sup n/4/)/sup /spl epsiv//). Thus the cost of the proposed parallel algorithm is O(2/sup n/), which is optimal, and an improved result over the past researches.

关键词： parallel algorithms Concurrent computing Cost function Algorithm design and analysis Computational modeling Computer aided instruction Computer science High performance computing Cryptography Read-write memory

来源：评论

学校读者我要写书评

暂无评论

An improved parallel thinning algorithm

An improved parallel thinning algorithm

引用

International Conference on Document Analysis and Recognition

作者： Lei Huang Genxun Wan Changping Liu Institute of Automation Chinese Academy and Sciences China

ISBN: (纸本)0769519601

This paper describes an improved thinning algorithm for binary images. We improve thinning algorithm from the fundamental properties such as connectivity, one-pixel width, robust to noise and speed. In addition, in order to overcome information loss, we integrated the contour and skeleton of pattern and proposed the threshold way. Some fundamental requirements of thinning and the shape of pattern are preserved very well. Algorithm is very robust to noise and eliminate some spurious branch. Above all, it can overcome the loss of information in pattern. Experimental results show the performance of the proposed algorithm.

关键词： Pixel Iterative algorithms Noise robustness Noise shaping Skeleton Shape parallel algorithms Automation Costs Character recognition

来源：评论

学校读者我要写书评

暂无评论

parallel mining of maximal frequent itemsets from databases

Parallel mining of maximal frequent itemsets from databases

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： S.M. Chung Congnan Luo Department of Computer Science and Engineering Wright State University Dayton OH USA

In this paper, we propose a parallel algorithm for mining maximal frequent itemsets from databases. A frequent itemset is maximal if none of its supersets is frequent. The new parallel algorithm is named parallel max-miner (PMM), and it is a parallel version of the sequential max-miner algorithm by R.J. Bayardo (1998). Most of existing mining algorithms discover the frequent k-itemsets on the kth pass over the databases, and then generate the candidate (k + 1)-itemsets for the next pass. Compared to those level-wise algorithms, PMM looks ahead at each pass and prunes more candidate itemsets by checking the frequencies of their supersets. We implemented PMM on a cluster of workstations, and evaluated its performance for various cases. PMM demonstrated better performance than other sequential and parallel algorithms, and its performance is quite scalable, even when there are large maximal frequent itemsets (i.e. long patterns) in databases.

关键词： Data mining Itemsets Association rules parallel algorithms Transaction databases Computer science Data engineering Clustering algorithms Frequency Workstations

来源：评论

学校读者我要写书评

暂无评论

Numerical performance of preconditioning techniques for the solution of complex sparse linear systems

引用

COMMUNICATIONS IN NUMERICAL METHODS IN ENGINEERING 2003年第1期19卷 37-48页

作者： Mazzia, A Pini, G Univ Padua Dipartimento Metodi & Modelli Matemat Sci Applica I-35131 Padua Italy

Preconditioning techniques based on ILU decomposition, on Frobenius norm minimization and on factorized sparse approximate inverse are considered. These algorithms are applied with conjugate gradient-type methods, namely Bi-CGSTAB, QMR and TFQMR for the solution of complex, large, sparse linear systems. The results of numerical experiments in scalar environment with matrices arising from transport in porous media, quantum chemistry, structural dynamics and electromagnetism are analysed. The preconditioner that appears most significant in parallel environment (based on factorized sparse approximate inverse) is then employed on a Cray T3E supercomputer. The experimental results show the satisfactory parallel performance of the proposed algorithm. Copyright (C) 2003 John Wiley Sons, Ltd.

关键词： complex sparse linear systems preconditioned iterative methods incomplete factorizations approximate inverses parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

Average case analysis-based protocols to initialize packet radio networks

引用

WIRELESS COMMUNICATIONS & MOBILE COMPUTING 2003年第4期3卷 539-548页

作者： Myoupo, LF Thimonier, L Ravelomanana, V Univ Picardie Fac Math & Informat LaRIA F-80000 Amiens France Univ Paris 13 Inst Galilee CNRS UPRES A 7030LIPN F-93430 Villetaneuse France

We propose two randomized protocols by which n (n not known) initially identical stations of a Packet Radio Network (PRN) are assigned ID numbers from 1 to n to distinguish them. They run regardless of the number of stations per channel. The first one is a naive protocol and is derived from recursive probabilistic divide-and-conquer techniques. It requires n/ln k broadcast rounds, where k is the number of communication channels. The second solution needs the well-known prefix sums algorithm and we show that in this scenario the described protocol terminates in 0(n/k) broadcast rounds on the average case whenever k less than or equal to n/ln n. These results are obtained by means of the average case analysis of algorithms, using probabilistic generating functions and formal methods. Surprisingly, our last protocol performs as well as the efficiency-oriented protocol of Hayashi et al. in [1,2], which depends on the number of stations per channel. And moreover, it can handle the case where k is an element of [n/3 Inn, n/ln n]. Copyright (C) 2003 John Wiley Sons, Ltd.

关键词： broadcast communication model wireless networks personal communication systems mobile computing parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

Last-generation applied artificial intelligence for energy management in building automation

引用

IFAC Proceedings Volumes 2003年第13期36卷 73-77页

作者： Yoseba K. Penya Institute of Computer Technology Vienna University of Technology Guβhausstraβ6 27-29 A1040 Vienna Austria

Artificial intelligence has devised solutions for scheduling problems that haven't been already applied to building-automation specific issues and could be beneficial. This paper presents one of these methods, a parallel-genetic algorithm for the optimisation of a multi-objective demand-side management system, and outlines the design of the architecture that supports the use of that algorithm.

关键词： Energy management systems genetic algorithms parallel algorithms scheduling algorithms automation artificial intelligence

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：