检索结果-内蒙古大学图书馆

2010 ACM/IEEE International Conference for high performance computing, Networking, Storage and Analysis, SC 2010

作者： Hamada, Tsuyoshi Nitadori, Keigo Nagasaki Advanced Computing Center Nagasaki University 1-14 Bunkyo-machi Nagasaki-city Nagasaki 852-8521 Japan RIKEN Advanced Science Institute High-Performance Molecular Simulation Team 61-1 Onocho Tsurumi Yokohama-city Kanagawa 230-0046 Japan

ISBN: (纸本)9781424475575

We present the results of a hierarchical N-body simulation on DEGIMA, a cluster of PCs with 576 graphic processing units (GPUs) and using an InfiniBand interconnect. DEGIMA stands for DEstination for GPU Intensive MAchine, and is located at Nagasaki Advanced computing center (NACC), Nagasaki University. In this work, we have upgraded DEGIMA's interconnect using InfiniBand. DEGIMA is composed by 144 nodes with 576 GT200 GPUs. An astrophysical N-body simulation with 3,278,982,596 particles using a treecode algorithm shows a sustained performance of 190.5 Tflops on DEGIMA. The overall cost of the hardware was $411,921 dollars. The maximum corrected performance is 104.8 Tflops for the simulation, resulting in a cost performance of 254.4 MFlops/$. This corrections is performed by counting the FLOPS based on the most efficient CPU algorithm. Any extra FLOPS that arise from the GPU implementation and parameter differences are not included in the 254.4 MFLOPS/$. © 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

simulation analysis of running safety on long span cable-stayed bridge based on parallel computing

引用

Zhendong yu Chongji/Journal of Vibration and Shock 2010年第7期29卷 5-8+78页

作者： Du, Xin-Guang Jin, Xian-Long Chen, Xiang-Dong High Performance Computing Center Shanghai Jiao Tong University Shanghai 200240 China State Key Laboratory of Mechanical System and Vibration Shanghai Jiao Tong University Shanghai 200240 China

Based on the Finite Element Analysis Method, the running safety of light rail train on long span double deck cable-stayed bridge is evaluated. The total element and nodes amount of the refined 3D couple vibration finite element model, which the bridge is made up of main truss, cable, tower, piers, cap, piles, and soil and the light rail vehicles include the car body, bogie, wheel-set, and spring-damp system, is over 2.2 million. The model is solved on Dawning 4000A supercomputer of Shanghai Supercomputer center based on the Contact Balance Bisection (CBB) algorithm for parallel computing and the vertical and transverse wheel-rail interaction force are obtained. According to the derailment coefficient and the wheel load reduction rate under maximum velocity, the running safety of light rail train is analyzed, which provides the reference for the safety design of long span cable-stayed bridge.

关键词： Supercomputers

来源：评论

学校读者我要写书评

暂无评论

Synchronizing the timestamps of concurrent events in traces of hybrid MPI/OpenMP applications

Synchronizing the timestamps of concurrent events in traces ...

引用

作者： Becker, Daniel Geimer, Markus Rabenseifner, Rolf Wolf, Felix German Research School for Simulation Sciences Laboratory for Parallel Programming 52062 Aachen Germany RWTH Aachen University Department of Computer Science 52056 Aachen Germany Forschungszentrum Jülich Jülich Supercomputing Centre 52428 Jülich Germany University of Stuttgart High Performance Computing-Center 70550 Stuttgart Germany

ISBN: (纸本)9780769542201

Event traces are helpful in understanding the performance behavior of parallel applications since they allow the indepth analysis of communication and synchronization patterns. However, the absence of synchronized clocks on most cluster systems may render the analysis ineffective because inaccurate relative event timings may misrepresent the logical event order and lead to errors when quantifying the impact of certain behaviors or confuse the users of time-line visualization tools by showing messages flowing backward in time. In our earlier work, we have developed a scalable algorithm that eliminates inconsistent inter-process timings postmortem in traces of pure MPI applications. Since hybrid programming, the combination of MPI and OpenMP in a single application, is becoming more popular on clusters in response to rising numbers of cores per chip and widening shared-memory nodes, we present an extended version of the algorithm that in addition to message-passing event semantics also preserves and restores shared-memory event semantics. © 2010 IEEE.

关键词： Message passing

来源：评论

学校读者我要写书评

暂无评论

Synchronizing the Timestamps of Concurrent Events in Traces of Hybrid MPI/OpenMP Applications

Synchronizing the Timestamps of Concurrent Events in Traces ...

引用

IEEE International Conference on Cluster computing

作者： Daniel Becker Markus Geimer Rolf Rabenseifner Felix Wolf Department of Computer Science RWTH Aachen University Aachen Germany German Research School for Simulation Sciences Laboratory for Parallel Programming Aachen Germany Forschungszentrum Jülich Jülich Supercomputing Centre Jülich Germany University of Stuttgart High Performance Computing-Center Stuttgart Germany

Event traces are helpful in understanding the performance behavior of parallel applications since they allow the in-depth analysis of communication and synchronization patterns. However, the absence of synchronized clocks on most cluster systems may render the analysis ineffective because inaccurate relative event timings may misrepresent the logical event order and lead to errors when quantifying the impact of certain behaviors or confuse the users of time-line visualization tools by showing messages flowing backward in time. In our earlier work, we have developed a scalable algorithm that eliminates inconsistent inter-process timings postmortem in traces of pure MPI applications. Since hybrid programming, the combination of MPI and OpenMP in a single application, is becoming more popular on clusters in response to rising numbers of cores per chip and widening shared-memory nodes, we present an extended version of the algorithm that in addition to message-passing event semantics also preserves and restores shared-memory event semantics.

关键词： Clocks Message systems Semantics Synchronization Algorithm design and analysis Interpolation

来源：评论

学校读者我要写书评

暂无评论

Numerical studies on micropart self-alignment using surface tension forces

引用

Microfluidics and Nanofluidics 2009年第1期6卷 63-75页

作者： Lin, Cheng Tseng, Fangang Kan, Heng-Chuan Chieng, Ching-Chang Department of Engineering and System Science National Tsing Hua University Kuang Fu Rd Hsinchu 30043 Taiwan National Center for High-Performance Computing Tainan County 74147 Taiwan

The Fluidic Self-Alignment Approach provides an alternative means for fast, economic, and precise handling of thousands of micro-scale parts. The present study aims to examine the important parameters which govern the mechanisms of the fluidic self-assembly process by numerical simulations. A simplified 2D model system consists of a solid plate, a micro-scale liquid slug and a micropart. The computational model is based on first principle conservation equations and is constructed by the coupling of two-phase modeling, solid structure modeling, and fluid-structure coupling. A matching experimental system is set up for the micropart of aspect ratio from 3:1 to 10:1 to validate the 2D computational simulations. simulations reveal that a high degree of hydrophilicity between the lubricant and the solid surfaces is required for the self-assembly of microparts. A lower lubricant height, a higher surface tension coefficient and a higher viscosity enforce the re-alignment/restoration process also. Characterization of the flow field inside lubricant slug also indicates that the asymmetry of the vortices/stress distribution at both ends of the lubricant meniscus is resulted as the micropart in a back-and-forth restoration process. © Springer-Verlag 2008.

关键词： Numerical models

来源：评论

学校读者我要写书评

暂无评论

Numerical simulation of bridge damage under blast loads

引用

WSEAS Transactions on Computers 2009年第9期8卷 1564-1574页

作者： Deng, Rong-Bing Jin, Xian-Long High Performance Computing Center School of Mechanical Engineering School of Mechanical Engineering No. 800 Dongchuan Road Shanghai 200240 China State Key Laboratory of Mechanical System and Vibration Shanghai Jiao Tong University School of Mechanical Engineering No. 800 Dongchuan Road Shanghai 200240 China

The numerical simulation of the structural damage of a steel truss bridge subjected to blast loading with the aid of a hydrocode is presented in this paper. A three-dimensional nonlinear finite element model of an actual bridge has been developed based on the drawing design of the Minpu II Bridge in Shanghai. The effects of mesh size on pressure distribution produced by explosions are also studied. Through the comparison between the calculation results and the experimental values, the reliability of the calculation is validated. All the process from the detonation of the explosive charge to deck crack, including the propagation of the blast wave and its interaction with the structure is reproduced. The numerical results show the damage of bridge parts and provide a global understanding of bridge under blast loads. It may be generated to supplement experimental studies for developing appropriate blast-resistant design guidelines for bridges in the future.

关键词： Trusses

来源：评论

学校读者我要写书评

暂无评论

An Application of the Mesh Generation and Refinement Tool to Mobile Bay, Alabama, USA

引用

AIP Conference Proceedings 2009年第1期1148卷 651-656页

作者： Wali Aziz Vladimir J. Alarcon William McAnally James Martin John Cartwright aComputational Simulation and Design Center High Performance Computing Collaboratory Mississippi State University Box 9627 Mississippi State MS 39762 USA bGeosystems Research Institute Northern Gulf Institute Mississippi State University Box 9627 Mississippi State MS 39762 USA cDepartment of Civil Engineering Mississippi State University Box 9546 Mississppi State MS 39762 USA

A grid generation tool, called the Mesh Generation and Refinement Tool (MGRT), has been developed using Qt4. Qt4 is a comprehensive C++ application framework which includes GUI and container class‐libraries and tools for cross‐platform development. MGRT is capable of using several types of algorithms for grid generation. This paper presents an application of the MGRT grid generation tool for creating an unstructured grid of Mobile Bay (Alabama, USA) that will be used for hydrodynamics modeling. The algorithm used in this particular application is the Advancing‐Front/Local‐Reconnection (AFLR) [1] [2]. This research shows results of grids created with MGRT and compares them to grids (for the same geographical container) created using other grid generation tools. The superior quality of the grids generated by MGRT is shown.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design and realization of a high productivity cluster-based network application reconfigurable accelerator board

引用

high Technology Letters 2008年第2期14卷 160-166页

作者：曾宇 Li Jun Sun Ninghui Wang Jie Liu Zhaohui National High Performance Computer Engineering Technology Center Beijing 100089 P.R. China National Research Center for Intelligent Computing System Institute of Computing TechnologyChinese Academy of Sciences Beijing 100190 P.R. Chinal Graduate School of Chinese Academy of Sciences Beijing 100039 P.R. China

Improving processor frequency to strengthen massive data processing capability will lead to incremen-tal server marginal costs and bring about a series of problems such as power consumption,managementcomplexity,*** on the field programmable gate array(FPGA),TCP offload engine(TOE),zero-copy and other key technologies,this paper describes the design and realization of a reconfigurable accel-erator *** this board,TCP/IP protocol will be moved to high-speed reconfigurable *** packets will be labeled according to the protocol and submitted to the upper data processingsoftware after IP-quintuple filtering in *** accelerator board obtains higher perfor-mance speed-up compared with ordinary NIC card.

关键词： high productivity reconfigurable accelerator

来源：评论

学校读者我要写书评

暂无评论

Anomaly detection and its application in grid

Anomaly detection and its application in grid

引用

2008 International Conference on Grid computing and Applications, GCA 2008

作者： Nguyen, Thuy T. Do, Thanh D. Nguyen, Dung T. Nguyen, Hiep C. Doan, Tung T. Trinh, Giang T. Department of Information System Hanoi University of Technology Hanoi Viet Nam High Performance Computing Center Hanoi University of Technology Hanoi Viet Nam Hanoi National University Hanoi Viet Nam

ISBN: (纸本)160132068X

Size and complexity of grid computing systems increase rapidly. Failures can have adverse effect on application executing on the system. Thus, failure information should be provided to the scheduler to make scheduling decisions that reduce effect of failures. In this paper, we survey anomaly detection methods and present in details window-based method together with experiments to evaluate it. The experiments show that window-based method is efficient in detecting anomalies. It may be implemented in a monitoring service. We introduce architecture of services to make failure information available for other services in grid. Using these services, grid schedulers and job/execution managers can get additional information for matching and handling task failures.

关键词： Scheduling

来源：评论

学校读者我要写书评

暂无评论

Replay-based synchronization of timestamps in event traces of massively parallel applications

Replay-based synchronization of timestamps in event traces o...

引用

37th International Conference on Parallel Processing Workshops, ICPP 2008

作者： Becker, Daniel Linford, John C. Rabenseifner, Rolf Wolf, Felix Forschungszentrum Jülich Institute for Advanced Simulation 52425 Jülich Germany RWTH Aachen University Department of Computer Science 52056 Aachen Germany Virginia Tech. Department of Computer Science Blacksburg VA 24061 United States University of Stuttgart High-Performance Computing-Center 70550 Stuttgart Germany

ISBN: (纸本)9780769533759

Event traces are helpful in understanding the performance behavior of message-passing applications since they allow in-depth analyses of communication and synchronization patterns. However, the absence of synchronized hardware clocks may render the analysis ineffective because inaccurate relative event timings can misrepresent the logical event order and lead to errors when quantifying the impact of certain behaviors. Although linear offset interpolation can restore consistency to some degree, inaccuracies and time-dependent drifts may still disarrange the original succession of events - especially during longer runs. In our earlier work, we have presented an algorithm that removes the remaining violations of the logical event order postmortem and, in addition, have outlined the initial design of a parallel version. Here, we complete the parallel design and describe its implementation within the SCALASCA trace-analysis framework. We demonstrate its suitability for large-scale applications running on more than thousand application processes and show how the correction can improve the trace analysis of a real-world application example. © 2008 IEEE.

关键词： Trace analysis

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：