检索结果-内蒙古大学图书馆

2010 IEEE World Congress on computational Intelligence

作者： Tsutsui, Shigeyoshi Fujimoto, Noriyuki Dept Management & Informat Sci 5-4-33 Amammihigashi Osaka 5808502 Japan Osaka Prefecture Univ Grad Sch Sci Osaka 5998531 Japan

ISBN: (纸本)9781424481262

This paper proposes an evolutionary algorithm for solving QAPs with parallel independent run using gpu computation and give sastatistical analysis on how speedup can be attained with this model. With the proposed model, we achieve a gpu computation performance that is nearly proportional to the number of equipped multi-processors (MPs) in the gpus. We explain these computational results by performing statistical analysis. Regarding performance comparison to CPU computations, gpu computation shows a speedup of x4.4 and x7.9 on average using a single gpu and two gpus, respectivey

关键词： gpu computation QAP coprocessors evolutionary algorithm evolutionary computation independent run multiprocessors parallel evolutionary computation parallel processing statistical analysis statistical analysis

来源：评论

学校读者我要写书评

暂无评论

An enhanced analytical-based geometry processor for the volume of solid (VoS) method in CPU and gpu computations

引用

COMPUTERS & FLUIDS 2025年 293卷

作者： Suprianto, Fandi D. Chern, Ming-Jyh Wang, Chin-Cheng Natl Taiwan Univ Sci & Technol Dept Mech Engn 43 Keelung RdSec 4 Taipei City 106335 Taiwan Natl Taipei Univ Technol Dept Vehicle Engn 1 Sec 3Zhongxiao E Rd Taipei City 106344 Taiwan

This paper introduces an analytical-based geometry processor which improves the efficiency of volume-of-solid (VoS) CFD solvers for fluid-structure interaction (FSI) scenarios with rigid body motion. The combination of the VoS method and the direct forcing immersed boundary (DFIB) method speeds up geometry creation while significantly reducing computational workload during unsteady simulations of solid structure movement or deformation. The proposed geometry processor's effectiveness is demonstrated in the vortex-induced vibrations (VIV) scenario of a single circular cylinder, where the geometry was dynamically modified at each timestep to correspond to the cylinder's passive motion. Comparisson studies show that the enhanced VoS function accelerates geometry construction process significantly faster than the ray-casting method while maintaining the same level of accuracy. Furthermore, its gpu implementation consistently achieves significant speedup across various computational loads, indicating superior scalability. This proposed method not only works well for simple geometries, but it also supports a wide range of multi-body curvilinear forms, including extruded spans and fully defined 3D geometries, making it suitable for a variety of applications.

关键词： Geometry processor Volume of solid function DFIB gpu computation

来源：评论

学校读者我要写书评

暂无评论

Efficient implementations of a Born Series for computing photoacoustic field from a collection of erythrocytes

引用

PHOTOACOUSTICS 2025年 43卷

作者： Mandal, Ujjal Singh, Navroop Singh, Kartikay Hagone, Vinit Nana Singh, Jagpreet Anand, Anshu S. Cox, Ben T. Saha, Ratan K. Indian Inst Informat Technol Allahabad Dept Appl Sci Prayagraj 211015 Uttar Pradesh India Indian Inst Technol Ropar Dept Comp Sci & Engn St 29 Ropar 140001 Punjab India Indian Inst Informat Technol Allahabad Dept Informat Technol Prayagraj 211015 Uttar Pradesh India UCL Dept Med Phys & Biomed Engn Gower St London WC1E 6BT England

Numerical implementation of the Born series procedure is a computationally expensive task. Various computational strategies have been adopted and tested in this work for fast execution of the convergent Born series (CBS) algorithm for solving inhomogeneous Helmholtz equation in the context of biomedical photoacoustics (PAs). The PA field estimated by the CBS method for a solid circular disk approximating a red blood cell exhibits excellent agreement with the analytical result. It is observed that PA pressure map for a collection of red blood cells (mimicking blood) retains the signature of multiple scattering of acoustic waves by the acoustically inhomogeneous PA sources. The developed numerical tool realizing the CBS algorithm compatible with systems having multiple graphics processing units can be utilized further for accurate and fast estimation of the PA field for large tissue media.

关键词： Helmholtz equation Green's function Convergent Born series Multiple scattering Multi-thread/parallel computation gpu computation

来源：评论

学校读者我要写书评

暂无评论

DistME: A Fast and Elastic Distributed Matrix computation Engine using gpus 19

DistME: A Fast and Elastic Distributed Matrix Computation En...

引用

ACM SIGMOD International Conference on Management of Data (SIGMOD)

作者： Han, Donghyoung Nam, Yoon-Min Lee, Jihye Park, Kyongseok Kim, Hyunwoo Kim, Min-Soo DGIST Daegu South Korea KISTI Daejeon South Korea

ISBN: (纸本)9781450356435

Matrix computation, in particular, matrix multiplication is time-consuming, but essentially and widely used in a large number of applications in science and industry. The existing distributed matrix multiplication methods only focus on either low communication cost (i.e., high performance) with the risk of out of memory or large-scale processing with high communication overhead. We propose a distributed elastic matrix multiplication method called CuboidMM that achieves both high performance and large-scale processing. We also propose a gpu acceleration method that can be combined with CuboidMM. CuboidMM partitions matrices into cuboids for optimizing the network communication cost with considering memory usage per task, and the gpu acceleration method partitions a cuboid into subcuboids for optimizing the PCI-E communication cost with considering gpu memory usage. We implement a fast and elastic matrix computation engine called DistME by integrating CuboidMM with gpu acceleration on top of Apache Spark. Through extensive experiments, we have demonstrated that CuboidMM and DistME significantly outperform the state-of-the-art methods and systems, respectively, in terms of both performance and data size.

关键词： Matrix multiplication Distributed data-parallel system gpu computation

来源：评论

学校读者我要写书评

暂无评论

computation of Fresnel holograms and diffraction-specific coherent panoramagrams for full-color holographic displays based on anisotropic leaky-mode modulators 29

Computation of Fresnel holograms and diffraction-specific co...

引用

Conference on Practical Holography XXIX - Materials and Applications

作者： Jolly, Sundeep Dreshaj, Ermal Bove, V. Michael, Jr. MIT Media Lab Cambridge MA 02139 USA

ISBN: (纸本)9781628414769

We have previously introduced a computational architecture suitable for driving full-color holographic display systems based around anisotropic leaky-mode modulators;this architecture appropriately handles single-sideband modulation and frequency-division multiplexing of spectral bands that correspond to the independent red, green, and blue color channels in the display output. In this paper, we describe an implementation for driving the MIT/BYU Mark IV holographic display system with such a computational approach, in cases of both pre-computed Fresnel CGHs and real-time, gpu-based diffraction-specific coherent panoramagrams. Real-time holographic images of nearly VGA-resolution (468 lines) are generated via three dual-head NVIDIA gpus via a CUDA-based implementation that encompasses the requisite orthographic view generation from 3-D data sources, parallel vector-based fringe computation per hogel and per color, single-sideband modulation, and frequency-division multiplexing. We present the first results of this scheme and review the resulting performance metrics.

关键词： dynamic holographic displays computational display holography acousto-optic modulation integrated optics single-sideband modulation wavelength-division multiplexing gpu computation

来源：评论

学校读者我要写书评

暂无评论

Establishment of 3D numerical wave flume and its application to the wave propagation based on SPH method

引用

OCEAN ENGINEERING 2024年 313卷

作者： Yang, Xi Liang, Guangqi Hu, Taian Zhang, Guiyong Zhang, Zhifan Dalian Univ Technol Sch Naval Architecture State Key Lab Struct Anal Optimizat & CAE Software Dalian 116024 Peoples R China Collaborat Innovat Ctr Adv Ship & Deep Sea Explora Shanghai 200240 Peoples R China Sun Yat Sen Univ Sch Ocean Engn & Technol Zhuhai 519082 Peoples R China Marine Design & Res Inst China Shanghai 200023 Peoples R China

Waves serve as the fundamental and pervasive driving force in the ocean. The interactions between marine structures and waves have consistently occupied a pivotal position in the field of ocean engineering. In this work, a three-dimensional (3D) numerical wave flume is constructed based on the smoothed particle hydrodynamics (SPH) as a complement to the physical tank to study wave propagation. A modified symmetric pressure gradient operator is employed to guarantee momentum conservation and surmount the dissipation. A 3D SPH procedure based on the gpu is developed, and a sponge layer is incorporated to reduce the computational resources while enhancing the computational efficiency. The developed 3D numerical wave tank successfully simulates the problems of regular wave propagation (first-order regular wave, second-order stokes wave), irregular wave propagation (solitary wave, freak wave), and wave-structure interaction (first-order regular wave passes breakwater, second-order stokes wave passes breakwater). The numerical results are in good agreement with the reference solutions, indicating that the current method is an effective and reliable approach for simulating complex wave problems.

关键词： SPH Wave propagation 3D numerical wave tank gpu computation

来源：评论

学校读者我要写书评

暂无评论

Parallel spectral element method for guided wave based structural health monitoring

引用

SMART MATERIALS AND STRUCTURES 2020年第9期29卷

作者： Kudela, P. Moll, J. Fiborek, P. Polish Acad Sci Inst Fluid Flow Machinery PL-80231 Gdansk Poland Goethe Univ Frankfurt Dept Phys D-60438 Frankfurt Germany

Parallel implementation of the spectral element method is developed in which flat shell spectral elements are utilized for spatial domain representation. The implementation is realised by using Matlab Parallel Computing Toolbox and optimized for Graphics Processing Unit (gpu) computation. In this way, considerable computation speed-up can be achieved in comparison to computation on conventional processors. The implementation includes an interpolation of wavefield on a uniform grid. The method was tested on experimental data set available on the Open Guided Waves platform. The qualitative comparison was performed on full wavefield data, whereas quantitative comparison was made directly on signals of propagating Lamb waves registered by piezoelectric transducers. In both cases, good agreement between numerical and experimental results was achieved. The proposed method is particularly useful for structural health monitoring algorithms in which signal parameters are required such as wave velocity dependence on the angle of propagation. Moreover, it enables model-assisted damage size estimation in which the damage influence curve is estimated based on large numerical data sets and sparse experimental data. Due to relatively short computation times, large data sets can be generated and used for machine learning or other soft computing methods opening up new possibilities in health monitoring of metallic and composite structures.

关键词： spectral element method guided waves parallel implementation delamination gpu computation Open Guided Waves

来源：评论

学校读者我要写书评

暂无评论

A high precision extrapolation method in multiphase-field model for simulating dendrite growth

引用

JOURNAL OF CRYSTAL GROWTH 2018年 490卷 25-34页

作者： Yang, Cong Xu, Qingyan Liu, Baicheng Tsinghua Univ Minist Educ Key Lab Adv Mat Proc Technol Sch Mat Sci & Engn Beijing 100084 Peoples R China

The phase-field method coupling with thermodynamic data has become a trend for predicting the microstructure formation in technical alloys. Nevertheless, the frequent access to thermodynamic database and calculation of local equilibrium conditions can be time intensive. The extrapolation methods, which are derived based on Taylor expansion, can provide approximation results with a high computational efficiency, and have been proven successful in applications. This paper presents a high precision second order extrapolation method for calculating the driving force in phase transformation. To obtain the phase compositions, different methods in solving the quasi-equilibrium condition are tested, and the M-slope approach is chosen for its best accuracy. The developed second order extrapolation method along with the M-slope approach and the first order extrapolation method are applied to simulate dendrite growth in a Ni-Al-Cr ternary alloy. The results of the extrapolation methods are compared with the exact solution with respect to the composition profile and dendrite tip position, which demonstrate the high precision and efficiency of the newly developed algorithm. To accelerate the phase-field and extrapolation computation, the graphic processing unit (gpu) based parallel computing scheme is developed. The application to large-scale simulation of multi-dendrite growth in an isothermal cross-section has demonstrated the ability of the developed gpu-accelerated second order extrapolation approach for multiphase-field model. (C) 2018 Elsevier B.V. All rights reserved.

关键词： Extrapolation method Phase-field Dendrite growth gpu computation

来源：评论

学校读者我要写书评

暂无评论

gpu-based parallel vertex substitution algorithm for the p-median problem

引用

COMPUTERS & INDUSTRIAL ENGINEERING 2013年第1期64卷 381-388页

作者： Lim, Gino J. Ma, Likang Univ Houston Dept Ind Engn Houston TX 77204 USA

We introduce a gpu-based parallel vertex substitution (pVS) algorithm for the p-median problem using the CODA architecture by NVIDIA. pVS is developed based on the best profit search algorithm, an implementation of vertex substitution (VS), that is shown to produce reliable solutions for p-median problems. In our approach, each candidate solution in the entire search space is allocated to a separate thread, rather than dividing the search space into parallel subsets. This strategy maximizes the usage of CPU parallel architecture and results in a significant speedup and robust solution quality. computationally, pVS reduces the worst case complexity from sequential VS's O(p . n(2)) to O(p . (n - p)) on each thread by parallelizing computational tasks on GPO implementation. We tested the performance of pVS on two sets of numerous test cases (including 40 network instances from OR-lib) and compared the results against a CPU-based sequential VS implementation. Our results show that pVS achieved a speed gain ranging from 10 to 57 times over the traditional VS in all test network instances. (C) 2012 Elsevier Ltd. All rights reserved.

关键词： Parallel computing Vertex substitution p-Median problem gpu computation

来源：评论

学校读者我要写书评

暂无评论

Ultrasonic imaging using signal post-processing for a flexible array transducer

引用

NDT & E INTERNATIONAL 2016年 82卷 13-25页

作者： Nakahata, K. Tokumasu, S. Sakai, A. Iwata, Y. Ohira, K. Ogura, Y. Ehime Univ Dept Civil & Environm Engn 3 Bunkyo Matsuyama Ehime 7908577 Japan Japan Probe Co Ltd Ctr Res & Dev Minami Ku 1-1-14 Nakamura Yokohama Kanagawa 2320033 Japan

In this study, we used a flexible array transducer to obtain images of flaws in materials with irregular surfaces. The transducer was manufactured from a 1-3 composite made of piezoelectric zirconate titanate and epoxy resin, and covered with neoprene rubber layers. An approach that involved combining a flexible array transducer and full matrix capture (FMC) is effective, because the surface geometry and flaw signal data can be separately measured and merged during post-processing. The disadvantage of the flexible array transducer is its narrow frequency band due to the thin damping material. To enhance the spatial resolution, we used the scattering amplitude extracted from raw signal data. We also introduced a numerical apodization technique to suppress the influence of side lobe. We validated the performance of the proposed method by measuring the signals from artificial flaws in aluminum specimens, performing high-speed FMC imaging using graphic processing unit computation. (C) 2016 Elsevier Ltd. All rights reserved.

关键词： Flaw imaging Scattering amplitude Flexible array transducer Full matrix capture Apodization gpu computation

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：