检索结果-内蒙古大学图书馆

2012 11th international symposium on parallel and distributed computing, ISPDC 2012

作者： Zumbusch, Gerhard Institut für Angewandte Mathematik Friedrich-Schiller-Universität Jena Jena Germany

ISBN: (纸本)9780769548050

Current CPU and GPU architectures heavily use data and instruction parallelism at different levels. Floating point operations are organised in vector instructions of increasing vector length. For reasons of performance it is mandatory to use the vector instructions efficiently. Several ways of tuning a model problem finite difference stencil computation are discussed. The combination of vectorisation and an interleaved data layout, cache aware algorithms, loop unrolling, parallelisation and parameter tuning lead to optimised implementations at a level of 90% peak performance of the floating point pipelines on recent Intel Sandy Bridge and AMD Bulldozer CPU cores, both with AVX vector instructions as well as on Nvidia Fermi/ Kepler GPU architectures. Furthermore, we present numbers for parallel multi-core/ multi-processor and multi-GPU configurations. They represent regularly more than an order of speed up compared to a standard implementation. The analysis may also explain deficiencies of automatic vectorisation for linear data layout and serve as a foundation of efficient implementations of more complex expressions. © 2012 IEEE.

关键词： Vectors

来源：评论

学校读者我要写书评

暂无评论

A distributed methodology for imbalanced classification problems

A distributed methodology for imbalanced classification prob...

引用

2012 11th international symposium on parallel and distributed computing, ISPDC 2012

作者： Lemnaru, Camelia Cuibus, Mihai Bona, Adrian Alic, Andrei Potolea, Rodica Computer Science Department Technical University of Cluj-Napoca Cluj-Napoca Romania

ISBN: (纸本)9780769548050

Current important challenges in data mining research are triggered by the need to address various particularities of real-world problems, such as imbalanced data and error cost distributions. This paper presents distributed Evolutionary Cost-Sensitive Balancing, a distributed methodology for dealing with imbalanced data and - if necessary - cost distributions. The method employs a genetic algorithm to search for an optimal cost matrix and base classifier settings, which are then employed by a cost-sensitive classifier, wrapped around the base classifier. Individual fitness computation is the most intensive task in the algorithm, but it also presents a high parallelization potential. Two different parallelization alternatives have been explored: a computation-driven approach, and a data-driven approach. Both have been developed within the Apache Watchmaker framework and deployed on Hadoop-based infrastructures. Experimental evaluations performed up to this point have indicated that the computation-driven approach achieves a good classification performance, but does not reduce the running time significantly, the data-driven approach reduces the running time for slow algorithms, such as the kNN and the SVM, while still yielding important performance improvements. © 2012 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Cross-stratum optimization aware provisioning algorithm for cloud data centers

Cross-stratum optimization aware provisioning algorithm for ...

引用

2012 10th IEEE international symposium on parallel and distributed Processing with Applications, ISPA 2012

作者： Tursunova, Shahnaza Lee, Tae-Ho Kodirov, Nodir Choi, Tae-Sang Yuseong-gu Daejeon 305-700 Korea Republic of

ISBN: (纸本)9780769547015

As virtualization technology matured, the concept of virtual network environment has emerged. Because of the flexibility of virtualization, the cloud computing services can serve various needs of its customers more efficiently. However, in some of virtual network provisioning algorithms decisions are made without any consideration of underlying network. In this paper, we are focusing on cross stratum optimization opportunities, e.g., combined optimization of resources in the application and network stratums, and proposing and an efficient and optimized provisioning algorithm. We propose a CSO-aware provisioning algorithm and show its benefits over traditional resource allocation. © 2012 IEEE.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Virtualizing general purpose GPUs for high performance cloud computing: An application to a fluid simulator

Virtualizing general purpose GPUs for high performance cloud...

引用

2012 10th IEEE international symposium on parallel and distributed Processing with Applications, ISPA 2012

作者： Di Lauro, Roberto Giannone, Flora Ambrosio, Luigia Montella, Raffaele Department of Applied Science University of Naples Parthenope Naples Italy

ISBN: (纸本)9780769547015

In this work we present an hypervisor-independent GPU Virtualization Service named GVirtus. It instantiates virtual machines able to access to the GPU in a transparent way. GPUs allow to speed up calculations over CPUs. Therefore, virtualizing GPUs is a major trend and can be considered a revolutionary tool for HPC. To test the performances of GVirtus we used a fluid simulator. Morover to exploit the computational power of GPUs in cloud computing we virtualized three different plugins for GVirtus Framework: Cuda Runtime, Cuda Driver and OpenCL plugins. Our results show that the overhead introduced by virtualization is almost irrelevant. © 2012 IEEE.

关键词： Virtualization

来源：评论

学校读者我要写书评

暂无评论

An efficient parallel implementation for three-dimensional incompressible pipe flow based on SIMPLE 12

An efficient parallel implementation for three-dimensional i...

引用

12th IEEE/ACM international symposium on Cluster, Cloud and Grid computing, CCGrid 2012

作者： Zhang, Ji-Lin Zhu, Li-Ting Wan, Jian Mao, Jie Xu, Xiang-Hua Jiang, Cong-Feng Di, Peng Department of Computer Science and Technology Hangzhou Dianzi University Hangzhou China School of Mechanical Engineering Hangzhou Dianzi University Hangzhou China School of Computer Science and Engineering University of New South Wales Sydney NSW Australia

ISBN: (纸本)9780769546919

SIMPLE (Semi-Implicit Method for Pressure-Linked Equations) algorithm is important in the simulation of steady flows. As the traditional 3-D SIMPLE algorithm is time-consuming, we propose a parallel SIMPLE algorithm based on a novel tiling strategy - alternate tiling, through replacing the original linear system and reordering the iteration space tiles. The novelty of our parallel algorithm lies in the introduction of the sequence of iteration space tiles as the sequence of execution, the time skewing technique to partition the iteration space, update operations of the grids from two directions alternately, and the improvement of the data locality. The effectiveness of the parallel algorithm and serial model of finite difference stencil algorithm are validated. Numerical experiments on distributed clusters show that the cache misses and the cost of communication and synchronization are reduced by reordering the tiles of iteration space, and the parallel SIMPLE algorithm based on alternate tiling has a good data locality and parallel efficiency in the three-dimensional incompressible pipe flow project. © 2012 IEEE.

关键词： parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

Virtual I/O forwarding for cloud-based HPC applications

Virtual I/O forwarding for cloud-based HPC applications

引用

2012 10th IEEE international symposium on parallel and distributed Processing with Applications, ISPA 2012

作者： Llopis, Pablo Martín, Gonzalo Bergua, Borja Carretero, Jesús University Carlos III Madrid Spain

ISBN: (纸本)9780769547015

Over the past years, the cloud, especially Infrastructure as a Service (IaaS) offerings, have seen a staggering rise in popularity due to the convenience and the economic pricing models. Even cloud service providers are starting to offer custom-sized HPC clusters as a service, with cloud-like pricing models. This can be of special interest to organizations who can not afford to acquire expensive HPC cluster hardware, or would require a different amount of computing power to meet computing demand for each project. From a user perspective, the cloud offers elastic and scalable computing power. Virtualization is a key ingredient to this advantage. However, HPC applications are often constrained by I/O, a problem which is made worse when running in a virtualized environment. In this work we present a flexible I/O virtualization solution, which is evaluated and compared to similar I/O forwarding technologies. Furthermore, we present a distributed I/O forwarding backend architecture which allows for complex, multi-host setups that can aim for different I/O strategies such as power proportionality and higher throughput. © 2012 IEEE.

关键词： computing power

来源：评论

学校读者我要写书评

暂无评论

SParTSim: A Space Partitioning Guided by Road Network for distributed Traffic Simulations 12

SParTSim: A Space Partitioning Guided by Road Network for Di...

引用

16th IEEE/ACM international symposium on distributed Simulation and Real Time Applications

作者： Ventresque, Anthony Bragard, Quentin Liu, Elvis S. Nowak, Dawid Murphy, Liam Theodoropoulos, Georgios Liu, Qi Univ Coll Dublin Lero UCD Sch Comp Sci & Informat Dublin Ireland Inst Natl Sci Appliquees Lyon Villeurbanne France Dublin Res Lab IBM Res Dublin Ireland IBM T J Watson Res Ctr Yorktown Hts NY USA

ISBN: (纸本)9780769548463

Traffic simulation can be very computationally intensive, especially for microscopic simulations of large urban areas (tens of thousands of road segments, hundreds of thousands of agents) and when real-time or better than real-time simulation is required. For instance, running a couple of what-if scenarios for road management authorities/police during a road incident: time is a hard constraint and the size of the simulation is relatively high. Hence the need for distributed simulations and for optimal space partitioning algorithms, ensuring an even distribution of the load and minimal communication between computing nodes. In this paper we describe a distributed version of SUMO, a simulator of urban mobility, and SParTSim, a space partitioning algorithm guided by road network for distributed simulations. It outperforms classical uniform space partitioning in terms of road segment cuts and load-balancing.

关键词： parallel and distributed Simulation Space Partitioning Graph Partitioning Road Network Traffic Simulation

来源：评论

学校读者我要写书评

暂无评论

Clustering based distributed phylogenetic tree construction

引用

EXPERT SYSTEMS WITH APPLICATIONS 2012年第1期39卷 89-98页

作者： Ruzgar, Esra Erciyes, Kayhan Izmir Univ Comp Eng Dept TR-35350 Izmir Turkey

Phylogenetic tree construction has received much attention recently due to the availability of vast biological data. In this study, we provide a three step method to build phylogenetic trees. Firstly, a density-based clustering algorithm is used to provide clusters of the population at hand using the distance matrix which shows the distances of the species. Secondly, a phylogenetic tree for each cluster is constructed by using the neighbor-joining (NJ) algorithm and finally, the roots of the small phylogenetic trees are connected again by the NJ algorithm to form one large phylogenetic tree. To our knowledge, this is the first method for building phylogenetic trees that uses clustering prior to forming the tree. As such, it provides independent phylogenetic tree formation within each cluster as the second step, hence is suitable for parallel/distributed processing, enabling fast processing of very large biological data sets. The proposed method, clustered neighbor-joining (CNJ) is applied to 145 samples from the Y-DNA Haplogroup G. Distances between male samples are the variation in their set of Y-chromosomal short tandem repeat (STR) values. We show that the clustering method we use is superior to other clustering methods as applied to Y-DNA data and also independent, fast distributed construction of phylogenetic trees is possible with this method. (C) 2011 Elsevier Ltd. All rights reserved.

关键词： Y-DNA Haplogroups Clustering Phylogenetic tree Neighbor-joining

来源：评论

学校读者我要写书评

暂无评论

Productive parallel linear algebra programming with unstructured topology adaption 12

Productive parallel linear algebra programming with unstruct...

引用

12th IEEE/ACM international symposium on Cluster, Cloud and Grid computing, CCGrid 2012

作者： Gottschling, Peter Hoefler, Torsten Technische Universität Dresden and SimuNova Helmholtzstr. 10 01069 Dresden Germany National Center for Supercomputing Applications University of Illinois at Urbana-Champaign 1205 W. Clark Street Urbana IL United States

ISBN: (纸本)9780769546919

Sparse linear algebra is a key component of many scientific computations such as computational fluid dynamics, mechanical engineering or the design of new materials to mention only a few. The discretization of complex geometries in unstructured meshes leads to sparse matrices with irregular patterns. Their distribution in turn results in irregular communication patterns within parallel operations. In this paper, we show how sparse linear algebra can be implemented effortless on distributed memory architectures. We demonstrate how simple it is to incorporate advanced partitioning, network topology mapping, and data migration techniques into parallel HPC programs by establishing novel abstractions. For this purpose, we developed a linear algebra library - parallel Matrix Template Library 4 - based on generic and meta-programming introducing a new paradigm: meta-tuning. The library establishes its own domain-specific language embedded in C++. The simplicity of software development is not paid by lower performance. Moreover, the incorporation of topology mapping demonstrated performance improvements up to 29%. © 2012 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Conservative distributed discrete event simulation on Amazon EC2 12

Conservative distributed discrete event simulation on Amazon...

引用

12th IEEE/ACM international symposium on Cluster, Cloud and Grid computing, CCGrid 2012

作者： Vanmechelen, Kurt De Munck, Silas Broeckhove, Jan Department of Mathematics and Computer Science University of Antwerp Antwerp Belgium

ISBN: (纸本)9780769546919

A discrete-event simulator's ability to distribute the execution of a simulation model allows one to deal with the memory limitations of a single computational resource, and thereby increase the scale or level of detail at which models can be studied. In addition, distribution has the potential to reduce the round trip time of a simulation by incorporating multiple computational cores into the simulation's execution. However, such gains can be voided by the overhead that time synchronization protocols introduce. These protocols are required to prevent the occurrence of causality errors during a parallel execution of a simulation. The overhead depends on the protocol, characteristics of the simulation model, and the architecture of the computational resources used. Recently, infrastructure-as-a-service offerings in cloud computing have introduced flexibility in acquiring computational resources on a pay-as-you-go basis. At present, it is unclear to what extent these offerings are suited for the distributed execution of discrete-event simulations, and how the characteristics of different resource types impact the runtime performance of distributed simulations. In this paper we investigate this issue, and assess the performance of different conservative time synchronization protocols on a range of cloud resource types that are currently available on Amazon EC2. © 2012 IEEE.

关键词： Discrete event simulation

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：