检索结果-内蒙古大学图书馆

作者： Watling, Robert P. Michigan Technological University

学位级别：M.S., Master of Science/Master of Surgery

Parallel graph processing is central to analytical computer science applications, and GPUs have proven to be an ideal platform for parallel graph processing. Existing GPU graph processing frameworks present performance improvements but often neglect two issues: the unpredictability of a given input graph and the energy consumption of the graph processing. Our prototype software, EEGraph (Energy Efficiency of Graph processing), is a flexible system consisting of several graph processing algorithms with configurable parameters for vertex update synchronization, vertex activation, and memory management along with a lightweight software-based GPU energy measurement scheme. We observe relationships between different configurations of our software, performance, and GPU energy for processing in-memory and out-of-memory graphs. The ideal parameters are discovered for specific input graphs by analyzing the observed relationships. We also present the utility of subgraph generation to predict the performance and energy consumption of complete graph configurations. EEGraph improves upon state-of-the-art GPU-based graph processing software by 2.08 times for performance and 1.60 times for GPU energy for processing in-memory graph datasets. Additionally, EEGraph improves upon the state-of-the-art by 3.30 times for performance and 1.63 times for GPU energy for processing large out-of-memory graph datasets.

关键词： Computer systems Energy graphical processing unit Graphs High performance computing Parallel programming

来源：评论

学校读者我要写书评

暂无评论

ORIN-3D-A new model for efficient simulation of landslide motion on a GPU using CUDA

引用

COMPUTERS AND GEOTECHNICS 2023年 153卷

作者： Aaron, Jordan Swiss Fed Inst Technol Geol Inst Chair Engn Geol Zurich Switzerland Swiss Fed Inst Forest Snow & Landscape Res WSL Birmensdorf Switzerland

Extremely-rapid flowlike landslide are a major hazard in many parts of the world, and managing their risk re-quires an understanding of the mechanisms that drive motion, as well as reliable predictions of their potential destructiveness. Numerical runout models are one tool that can be used for both these applications, however their utility is presently limited by their computational runtime. In the present work, a new depth-averaged, smooth particle hydrodynamics based model is implemented to run on a graphical processing unit. The new implementation provides a speedup of over two orders of magnitude, compared to a commonly used CPU based runout model. The new model has been validated, and then is used to back-analyse the Johnsons Landing Landslide. It is shown that increasing model resolution results in an accurate simulation of complex topographic interactions between the flowing landslide and the surface topography. The new model runs on commercially available GPU's, and should therefore be useful for both researchers and practitioners seeking to understand and quantify landslide hazard and risk.

关键词： Landslide runout modelling Rock avalanche Debris avalanche Smooth particle hydrodynamics graphical processing unit CUDA

来源：评论

学校读者我要写书评

暂无评论

Accelerated GPU based SPECT Monte Carlo simulations

引用

PHYSICS IN MEDICINE AND BIOLOGY 2016年第11期61卷 4001-4018页

作者： Garcia, Marie-Paule Bert, Julien Benoit, Didier Bardies, Manuel Visvikis, Dimitris CHRU Brest INSERM UMR 1101 LaTIM Brest France Ctr Rech Cancerol Toulouse UPS INSERM UMR 1037 Toulouse France

Monte Carlo (MC) modelling is widely used in the field of single photon emission computed tomography (SPECT) as it is a reliable technique to simulate very high quality scans. This technique provides very accurate modelling of the radiation transport and particle interactions in a heterogeneous medium. Various MC codes exist for nuclear medicine imaging simulations. Recently, new strategies exploiting the computing capabilities of graphical processing units (GPU) have been proposed. This work aims at evaluating the accuracy of such GPU implementation strategies in comparison to standard MC codes in the context of SPECT imaging. GATE was considered the reference MC toolkit and used to evaluate the performance of newly developed GPU Geant4-based Monte Carlo simulation (GGEMS) modules for SPECT imaging. Radioisotopes with different photon energies were used with these various CPU and GPU Geant4-based MC codes in order to assess the best strategy for each configuration. Three different isotopes were considered: Tc-99m, In-111 and I-131, using a low energy high resolution (LEHR) collimator, a medium energy general purpose (MEGP) collimator and a high energy general purpose (HEGP) collimator respectively. Point source, uniform source, cylindrical phantom and anthropomorphic phantom acquisitions were simulated using a model of the GE infinia II 3/8 '' gamma camera. Both simulation platforms yielded a similar system sensitivity and image statistical quality for the various combinations. The overall acceleration factor between GATE and GGEMS platform derived from the same cylindrical phantom acquisition was between 18 and 27 for the different radioisotopes. Besides, a full MC simulation using an anthropomorphic phantom showed the full potential of the GGEMS platform, with a resulting acceleration factor up to 71. The good agreement with reference codes and the acceleration factors obtained support the use of GPU implementation strategies for improving computational efficienc

关键词： Monte Carlo simulation single photon emission computed tomography graphical processing unit

来源：评论

学校读者我要写书评

暂无评论

GPU acceleration and data fitting: Agent-based models of viral infections can now be parameterized in hours

引用

JOURNAL OF COMPUTATIONAL SCIENCE 2022年第0期61卷

作者： Fain, Baylor G. Dobrovolny, Hana M. Texas Christian Univ Dept Phys & Astron Ft Worth TX 76129 USA

For many years, infectious disease modelers have used agent-based models to study the spread of viruses, but the models were too computationally intensive to fully replicate even in vitro experiments. Now, with technological advancements and accessible software, agent-based models can be used to their full potential. In this work, we created an agent-based model that expresses viral transmission and diffusion, can manipulate and track individual cells, and can be fit to real experimental data in a timely manner due to acceleration of computation with graphics processing units (GPUs). The use of GPUs allows simulations to run on desktop computers in a few minutes or hours, while still simulating an accurate number of cells to replicate infection experiments. This model can now be used to study in-host infections quickly, so that in the event of an outbreak or epidemic a treatment plan and course of action can be developed in less time.

关键词： Agent-based model Viral infection graphical processing unit Computer simulation

来源：评论

学校读者我要写书评

暂无评论

BIO-MIMETIC CLASSIFICATION ON MODERN PARALLEL HARDWARE: REALIZATIONS ON NVIDIAG® CUDA™

引用

INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL 2011年第7A期7卷 3825-3837页

作者： Nowotny, Thomas Muezzinoglu, Mehmet K. Huerta, Ramon Univ Sussex Sch Informat Brighton BN1 9QJ E Sussex England Univ Calif San Diego BioCircuits Inst La Jolla CA 92093 USA

Both the brain and modern digital architectures rely on massive parallelism for efficient solutions to demanding computational tasks, such as pattern recognition. In this paper, we implement a parallel classification scheme inspired by the insect brain in, two popular parallel computing frameworks, namely as an NVidia (R) CUDA (TM) implementation on a Tesla (TM) device and a brute force OpenMP (TM) parallel implementation on a quad-core CPU. When evaluating the systems on the MNIST data-set of handwritten digits, we can report that, compared with a standard serial implementation on a single CPU core, CUDA (TM) implementations of the bio-inspired classification provide a 7-to-11 fold speed-up, whereas the OpenMP (TM) implementation is 2-to-.4 times faster. Our results are a proof of concept that suggests that modern parallel computing architectures and bio-mimetic algorithms are compatible and that the CUDA (TM) solution on an NVidia Tesla (TM) C870 device at the time of writing has a small edge over an OpenMP solution on a recent quad core processor (3 GHz AMD (R) Phenom (TM) II X4 940).

关键词： Parallel hardware graphical processing unit CUDA OpenMP Bio-mimetic systems Classification Insect brain Olfactory system MNIST data set Multi-layer perceptron Random connections

来源：评论

学校读者我要写书评

暂无评论

Lagrangian computational fluid dynamics for nuclear Thermal-Hydraulics & safety

引用

NUCLEAR ENGINEERING AND DESIGN 2023年第1期405卷

作者： Jo, Young Beom Park, So-Hyun Kim, Eung Soo Seoul Natl Univ Dept Nucl Engn 1 Gwanak Ro Seoul South Korea Korea Atom Energy Res Inst 989-111 Daeduk Daero Daejeon 305353 South Korea

In recent years, Lagrangian numerical methods have gained considerable attention from various science & en-gineering fields with the rapid advancement of the computational environment. In the Lagrangian methods, the domain of interest is represented by a finite number of particles that carry individual properties and variables by solving discretized equations over the adjacent particles. The Lagrangian CFD method has a clear advantage over traditional grid-based numerical methods when it comes to handling free surfaces, multi-phase systems, and boundaries that undergo complex deformation. The Lagrangian CFD method possesses a notable advantage due to its Lagrangian characteristics, wherein the interface at the free surface and multi-phase boundary is imposed and determined by the motion of particles without additional efforts or manipulation for interface tracking. Thus, there are no constraints on the geometry of the fluid system. For these reasons, it is relatively easy to implement various types of physics, and hence, it is expected to fit well into the nuclear thermal-hydraulics & safety problems that traditional grid-based methods have intrinsic difficulties in managing. Despite these potentials originating from the numerical scheme, the young Lagrangian CFD methods still require further improvement in various aspects. More efforts should be made to reduce the computational cost, and many improvements are required in terms of the accuracy, stability, and adaptivity of Lagrangian approaches. It is also necessary to establish the robustness of various physical models for practical engineering applications. This paper briefly summarizes the recent progress of Lagrangian CFD methods focusing on smoothed particle hydrodynamics (SPH), which is the most widely used Lagrangian numerical method, and discusses its potential and challenges for nuclear thermal-hydraulics and safety applications.

关键词： Nuclear Thermal-Hydraulics Nuclear Safety Advanced Modeling High Performance Computing graphical processing unit Smoothed Particle Hydrodynamics Multi-physics

来源：评论

学校读者我要写书评

暂无评论

Monte Carlo lookup table-based inverse model for extracting optical properties from tissue-simulating phantoms using diffuse reflectance spectroscopy

引用

JOURNAL OF BIOMEDICAL OPTICS 2013年第3期18卷 037003-037003页

作者： Hennessy, Ricky Lim, Sam L. Markey, Mia K. Tunnell, James W. Univ Texas Austin Austin TX 78712 USA Univ Texas MD Anderson Canc Ctr Houston TX 77030 USA

We present a Monte Carlo lookup table (MCLUT)-based inverse model for extracting optical properties from tissue-simulating phantoms. This model is valid for close source-detector separation and highly absorbing tissues. The MCLUT is based entirely on Monte Carlo simulation, which was implemented using a graphics processing unit. We used tissue-simulating phantoms to determine the accuracy of the MCLUT inverse model. Our results show strong agreement between extracted and expected optical properties, with errors rate of 1.74% for extracted reduced scattering values, 0.74% for extracted absorption values, and 2.42% for extracted hemoglobin concentration values. (C) The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication

关键词： optical properties diffuse reflectance spectroscopy graphical processing unit lookup table inverse model

来源：评论

学校读者我要写书评

暂无评论

Elviz - exploration of metagenome assemblies with an interactive visualization tool

引用

BMC BIOINFORMATICS 2015年第1期16卷 1-8页

作者： Cantor, Michael Nordberg, Henrik Smirnova, Tatyana Hess, Matthias Tringe, Susannah Dubchak, Inna Joint Genome Inst Dept Energy Walnut Creek CA 94598 USA Univ Calif Davis Davis CA 95616 USA

Background: Metagenomics, the sequencing of DNA collected from an entire microbial community, enables the study of natural microbial consortia in their native habitats. Metagenomics studies produce huge volumes of data, including both the sequences themselves and metadata describing their abundance, assembly, predicted functional characteristics and environmental parameters. The ability to explore these data visually is critically important to meaningful biological interpretation. Current genomics applications cannot effectively integrate sequence data, assembly metadata, and annotation to support both genome and community-level inquiry. Results: Elviz (Environmental Laboratory Visualization) is an interactive web-based tool for the visual exploration of assembled metagenomes and their complex metadata. Elviz allows scientists to navigate metagenome assemblies across multiple dimensions and scales, plotting parameters such as GC content, relative abundance, phylogenetic affiliation and assembled contig length. Furthermore Elviz enables interactive exploration using real-time plot navigation, search, filters, axis selection, and the ability to drill from a whole-community profile down to individual gene annotations. Thus scientists engage in a rapid feedback loop of visual pattern identification, hypothesis generation, and hypothesis testing. Conclusions: Compared to the current alternative of generating a succession of static figures, Elviz can greatly accelerate the speed of metagenome analysis. Elviz can be used to explore both user-submitted datasets and numerous metagenome studies publicly available at the Joint Genome Institute (JGI). Elviz is freely available at http://***/viz and runs on most current web-browsers.

关键词： graphical processing unit Joint Genome Institute Metagenome Dataset Anaerobic Methane Oxidation Metagenome Assembly

来源：评论

学校读者我要写书评

暂无评论

Utilizing graphics processing unit to accelerate drug-symptom association mining

引用

COMPUTERS & ELECTRICAL ENGINEERING 2020年 86卷 106704-106704页

作者： Tian, Yun Scholer, Jesse Ji, Yanqing Rogers, Uri Shen, Fangyang Eastern Washington Univ Dept Comp Sci Cheney WA 99004 USA Ciena Telecomm Spokane WA USA Gonzaga Univ Dept Elect & Comp Engn Spokane WA USA Eastern Washington Univ Dept Engn & Design Cheney WA 99004 USA New York City Coll Tech Brooklyn NY USA

A limited number of graphics processing unit algorithms exist for frequent itemset and association rule mining. This paper attempts to address that gap by introducing algorithms that lend themselves to massively parallel processing in a tool we call GPUMiner. The performance of GPUMiner will be contrasted against classic algorithms developed for a central processing unit type architecture. Multiple optimizations are adopted to improve efficiency in our design, including separate bitmaps for drugs and symptoms, parallel reduction for sum operation and a thread combination matrix that enables multiple-drug combinations to explored. Experiments, using the popular test dataset ***, show that our GPUMiner is able to achieve a speedup of 13.7 in comparison to the existing implementation. In addition, we apply GPUMiner in discovering drug-symptom associations and report on some well-known symptoms associated with a single drug or a combination of multiple drugs. Published by Elsevier Ltd.

关键词： Frequent Itemset Mining Association Rule Mining graphical processing unit Drug-symptom Association Mining

来源：评论

学校读者我要写书评

暂无评论

Dual Channel Multiplier for First-Order Piecewise Approximation for GPU

引用

JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS 2022年第17期31卷 2250298-2250298页

作者： Ellaithy, Dina M. Elect Res Inst ERI Microelect Dept Cairo 12622 Egypt

This paper demonstrates an energy-efficient implementation of piecewise polynomial approximation to evaluate the complex functions of graphical processing units (GPUs). A novel approach is employed to implement the first-order piecewise approximation serially which leads to high savings in power and area. Dual channel multiplier (DCM) scheme is proposed to simplify the hardware architecture of the piecewise polynomial approximation. The proposed methodology is implemented with 90 nm CMOS technology and it can perform different complex functions using a simple multiplier hardware structure. DCM achieves improvement in energy saving by up to 81% at a penalty cost of 10 clock cycles. Simulation results confirm that this work attains at least 83%, and 55% saving in power and area as compared to the traditional techniques, respectively. The proposed methodology can implement any degree of the piecewise polynomial approximation utilizing the DCM.

关键词： Dual channel multiplier graphical processing unit low power piecewise polynomial approximation serial multipliers

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：