检索结果-内蒙古大学图书馆

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Karlsson, Christer Chen, Zizhong Colorado School of Mines Golden CO United States

ISBN: (纸本)9781424465347

A consequence of the fact that the number of processors in High Performance Computers (HPC) continues to increase is demonstrated by the correlation between Mean-Time-To-Failure(TMTTF ) and application execution time. The TMTTF is becoming shorter than the expected execution time for many next generation HPC applications. There is an ability to handle failure without a system-wide breakdown in most architecture, but many of the applications do not have a built-in ability to survive node failures. The purpose of this paper is to present an approach to develop a highly scalable technique to allow the next generation applications to survive node and/or link failure without aborting the computation. We will develop several strategies to improve the scalability of diskless checkpointing. The technique is scalable in the sense that when the number of processes increases, the overhead to handle k failures on p processes should remain as constant as possible. We will present the proposed technique, initial results together with remaining objectives and challenges. © 2010 ieee.

关键词： distributed computer systems

来源：评论

学校读者我要写书评

暂无评论

Characterizing energy efficiency of I/O intensive parallel applications on power-aware clusters

Characterizing energy efficiency of I/O intensive parallel a...

引用

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Ge, Rong Feng, Xizhou Subramanya, Sindhu Sun, Xian-He Dept. of Mathematics Statistics and Computer Science Marquette University Milwaukee WI 53201 United States Dept. of Computer Science Illinois Institute of Technology Chicago IL 60616 United States

ISBN: (纸本)9781424465347

Energy efficiency and parallel I/O performance have become two critical measures in high performance computing (HPC). However, there is little empirical data that characterize the energy-performance behaviors of parallel I/O workload. In this paper, we present a methodology to profile the performance, energy, and energy efficiency of parallel I/O access patterns and report our findings on the impacting factors of parallel I/O energy efficiency. Our study shows that choosing the right buffer size can change the energyperformance efficiency by up to 30 times. High spatial and temporal spacing can also lead to significant improvement in energy-performance efficiency (about 2X). We observe CPU frequency has a more complex impact, depending on the IO operations, spatial and temporal, and memory buffer size. The presented methodology and findings are useful for evaluating the energy efficiency of I/O intensive applications and for providing a guideline to develop energy efficient parallel I/O technology. © 2010 ieee.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Diagnosing permanent faults in distributed and parallel computing systems using artificial neural networks

Diagnosing permanent faults in distributed and parallel comp...

引用

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Elhadef, Mourad College of Engineering and Computer Science Abu Dhabi University Abu Dhabi United Arab Emirates

ISBN: (纸本)9781424465347

This paper deals with the problem of identifying faulty nodes (or units) in diagnosable distributed and parallel systems under the PMC model. In this model, each unit is tested by a subset of the other units, and it is assumed that, at most, a bounded subset of these units is permanently faulty. When performing testing, faulty units can incorrectly claim that fault-free units are faulty or that faulty units are faultfree. Since the introduction of the PMC model, significant progress has been made in both theory and practice associated with the original model and its offshoots. Nevertheless, this problem of efficiently identifying the set of faulty units of a diagnosable system remained an outstanding research issue. In this paper, we describe a new neural-network-based diagnosis algorithm, which exploits the off-line learning phase of artificial neural network to speed up the diagnosis algorithm. The novel approach has been implemented and evaluated using randomly generated diagnosable systems. The simulation results showed that the new neural-network-based fault identification approach constitutes an addition to existing diagnosis algorithms. Extreme faulty situations, where the number of faults is around the bound t, and large diagnosable systems have been also experimented to show the efficiency of the new neural-networkbased diagnosis algorithm. © 2010 ieee.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

A parallel particle swarm optimization algorithm for option pricing

A parallel particle swarm optimization algorithm for option ...

引用

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Prasain, Hari Jha, Girish Kumar Thulasiraman, Parimala Thulasiram, Ruppa Department of Computer Science University of Manitoba Winnipeg MB Canada

ISBN: (纸本)9781424465347

Option pricing is one of the challenging problems of computational finance. Nature-inspired algorithms have gained prominence in real world optimization problems such as in mobile ad hoc networks. The option pricing problem fits very well into this category of problems due to the ad hoc nature of the market. Particle swarm optimization (PSO) is one of the novel global search algorithms based on swarm intelligence. We first show that PSO could be effectively used for the option pricing problem. The results are compared with standard classical Black-Scholes-Merton model for simple European options. In this study, we developed two algorithms for option pricing using Particle Swarm Optimization (PSO). The first algorithm we developed is synchronous option pricing algorithm using PSO (SPSO), and the second is parallel synchronous option pricing algorithm. The pricing results of these algorithms are close when compared with classical Black-Scholes-Merton model for simple European options. We test our parallel Synchronous PSO algorithm in three architectures: shared memory machine using OpenMP, distributed memory machine using MPI and on a homogeneous multicore architecture running MPI and OpenMP (hybrid model). The results show that the hybrid model handles the load well as we increase the number of particles in simulation while maintaining equivalent accuracy. © 2010 ieee.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

Tackling consistency issues for runtime updating distributed systems

Tackling consistency issues for runtime updating distributed...

引用

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Bannò, Filippo Marletta, Daniele Pappalardo, Giuseppe Tramontana, Emiliano Scuola Superiore di Catania Università di Catania Italy Dipartimento di Matematica e Informatica Università di Catania Italy

ISBN: (纸本)9781424465347

Dynamic update capabilities allow a system to change some of its components without stopping execution, so as to cater for versioning and self-optimisation. In this paper, we propose FREJA to support transparent dynamic updates of classes of a distributed Java system. In doing so, FREJA purports to (i) ensure the consistency of the execution flow, and (ii) avoid that faults typical of a distributed environment interfere with the updating operations. FREJA's operations are transparent with respect to the system to be updated, thanks to appropriate selective bytecode rewriting of classes, performed at load-time. Consistency of the execution flow, for any state of the system to be updated and even in the face of network faults, is ensured by an ad-hoc protocol which checks the validity of changes and, should a fault manifest itself, falls back to a previous configuration. Dynamic updates, executed while safeguarding consistency, foster systems properties such as availability, maintainability and fault-tolerance, since by changing some parts of a running system bugs fixes and fault recovery can be performed. Hence, the proposed solution provides developers with valuable mechanisms which are paramount to obtain dependable systems. © 2010 ieee.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

parallel extraction and simplification of large isosurfaces using an extended tandem algorithm

引用

COMPUTER-AIDED DESIGN 2010年第2期42卷 129-138页

作者： Dupuy, Guilhem Jobard, Bruno Guillon, Sebastien Keskes, Noomane Komatitsch, Dimitri Univ Pau Comp Sci Lab LIUPPA Pau France Univ Pau Geophys Lab MIGP Pau France

In order to deal with the common trend in size increase of volumetric datasets, in the past few years research in isosurface extraction has focused on related aspects such as surface simplification and load-balanced parallel algorithms. We present a parallel, block-wise extension of the tandem algorithm [Attali D, Cohen-Steiner D, Edelsbrunner H. Extraction and simplification of iso-surfaces in tandem. In: SGP '05: proceedings of the third Eurographics symposium on Geometry processing. Aire-la-Ville, Switzerland: Eurographics Association;2005. p. 139-148], which simplifies on the fly an isosurface being extracted. Our approach minimizes the overall memory consumption using an adequate block splitting and merging strategy along with the introduction of a component dumping mechanism that drastically reduces the amount of memory needed for particular datasets such as those encountered in geophysics. As soon as detected, surface components are migrated to the disk along with a meta-data index (oriented bounding box, volume, etc) that permits further improved exploration scenarios (small component removal or particularly oriented component selection for instance). For ease of implementation, we carefully describe a master and worker algorithm architecture that clearly separates the four required basic tasks. We show several results Of Our parallel algorithm applied on a geophysical dataset of size 7000 x 1600 x 2000. (C) 2009 Elsevier Ltd. All rights reserved.

关键词： Geometric computation distributed design and geo-scientific applications

来源：评论

学校读者我要写书评

暂无评论

An adaptive I/O load distribution scheme for distributed systems

An adaptive I/O load distribution scheme for distributed sys...

引用

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Chen, Xin Langston, Jeremy He, Xubin Mao, Fengjiang Department of Electrical and Computer Engineering Tennessee Technological University Cookeville TN 38505 United States Department of Electronic and Information Engineering Shenzhen Polytechnic Shenzhen Guangdong 518055 China

ISBN: (纸本)9781424465347

A fundamental issue in a large-scale distributed system consisting of heterogeneous machines which vary in both I/O and computing capabilities is to distribute workloads with respect to the capabilities of each node to achieve the optimal performance. However, node capabilities are often not stable due to various factors. Simply using a static workload distribution scheme may not well match the capability of each node. To address this issue, we distribute workload adaptively to the change of system node capability. In this paper we present an adaptive I/O load distribution scheme to dynamically capture the I/O capabilities among system nodes and to predictively determine an suitable load distribution pattern. A case study is conducted by applying our load distribution scheme into a popular distributed file system PVFS2. Experiments results show that our adaptive load distribution scheme can dramatically improve the performance: up to 70% performance gain for writes and 80% for reads, and up to 63% overall performance loss can be avoided in the presence of an unstable Object Storage Device (OSD). © 2010 ieee.

关键词： Virtual storage

来源：评论

学校读者我要写书评

暂无评论

Efficient parallel algorithms for maximum-density segment problem

Efficient parallel algorithms for maximum-density segment pr...

引用

24th ieee International parallel and distributed processing symposium, IPDPS 2010

作者： Wang, Xue Qiu, Fasheng Prasad, Sushil K. Chen, Guantao Computer Science Georgia State University Atlanta United States Mathematics and Statistics Georgia State University Atlanta United States

ISBN: (纸本)9781424464432

One of the fundamental problems involving DNA sequences is to find high density segments of certain widths, for example, those regions with intensive guanine and cytosine (GC). Formally, given a sequence, each element of which has a value and a width, the maximum-density segment problem asks for the segment with the maximum density while satisfying minimum and possibly maximum width constraints. While several linear-time sequential algorithms have emerged recently due to its primitive-like utility, to our knowledge, no nontrivial parallel algorithm has yet been proposed for this topical problem. In this paper, we propose an 0(log2 n)-time CREW PRAM algorithm using n processors to solve the generalized maximum-density problem, with a minimum width constraint and non-uniform widths. Besides, we describe an efficient implementation of the parallel algorithm on manycore GPUs (nVIDIA GeForce GTX 280), taking advantage of the full programmability of CUDA. This algorithm can process up to million-size sequence within a second using an nVIDIA GeForce GTX 280, thus demonstrating the practicality of this algorithm as a basic primitive for scientists. This may also indicate suitability of modern GPU architectures as implementation platform for certain PRAM algorithms. © 2010 ieee.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

OpenCL - An effective programming model for data parallel computations at the Cell Broadband Engine

OpenCL - An effective programming model for data parallel co...

引用

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Breitbart, Jens Fohry, Claudia Research Group Programming Languages / Methodologies Universität Kassel Germany

ISBN: (纸本)9781424465347

Current processor architectures are diverse and heterogeneous. Examples include multicore chips, GPUs and the Cell Broadband Engine (CBE). The recent Open Compute Language (OpenCL) standard aims at efficiency and portability. This paper explores its efficiency when implemented on the CBE, without using CBE-specific features such as explicit asynchronous memory transfers. We based our experiments on two applications: matrix multiplication, and the client side of the Einstein@Home distributed computing project. Both were programmed in OpenCL, and then translated to the CBE. For matrix multiplication, we deployed different levels of OpenCL performance optimization, and observed that they pay off on the CBE. For the Einstein@Home application, our translated OpenCL version achieves almost the same speed as a native CBE version. Another main contribution of the paper is a proposal for an additional memory level in OpenCL, called static local memory. With little programming expense, it can lead to significant speedups such as factor seven for reduction. Finally, we studied two versions of the OpenCL to CBE mapping, in which the PPE component of the CBE does or does not take the role of a compute unit. © 2010 ieee.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

pFANGS: parallel high speed sequence mapping for next generation 454-roche sequencing reads

pFANGS: Parallel high speed sequence mapping for next genera...

引用

2010 ieee International symposium on parallel and distributed processing, Workshops and Phd Forum, IPDPSW 2010

作者： Misra, Sanchit Narayanan, Ramanathan Liao, Wei-Keng Choudhary, Alok Lin, Simon Department of Electrical Engineering and Computer Science Northwestern University Evanston IL 60208 United States Robert H. Lurie Comprehensive Cancer Center Northwestern University Chicago IL 60611 United States

ISBN: (纸本)9781424465347

Millions of DNA sequences (reads) are generated by Next Generation Sequencing machines everyday. There is a need for high performance algorithms to map these sequences to the reference genome to identify single nucleotide polymorphisms or rare transcripts to fulfill the dream of personalized medicine. In this paper, we present a high-throughput parallel sequence mapping program pFANGS. pFANGS is designed to find all the matches of a query sequence in the reference genome tolerating a large number of mismatches or insertions/ deletions. pFANGS partitions the computational workload and data among all the processes and employs loadbalancing mechanisms to ensure better process efficiency. Our experiments show that, with 512 processors, we are able to map approximately 31 million 454/Roche queries of length 500 each to a reference human genome per hour allowing 5 mismatches or insertion/deletions at full sensitivity. We also report and compare the performance results of two alternative parallel implementations of pFANGS: a shared memory OpenMP implementation and a MPI-OpenMP hybrid implementation.© 2010 ieee.

关键词： DNA sequences

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：