检索结果-内蒙古大学图书馆

HTMoL: full-stack solution for remote access, visualization, and analysis of molecular dynamics trajectory data

JOURNAL OF COMPUTER-AIDED MOLECULAR DESIGN 2018年第8期32卷 869-876页

作者： Carrillo-Tripp, Mauricio Alvarez-Rivera, Leonardo Israel Lara-Ramirez, Omar Javier Becerra-Toledo, Francisco Vega-Ramirez, Adan Quijas-Valades, Emmanuel Gonzalez-Zavala, Eduardo Cesar Gonzalez-Vazquez, Julio Garcia-Vieyra, Javier Beatriz Santoyo-Rivera, Nelly Victor Chapa-Vergara, Sergio Meneses-Viveros, Amilcar Inst Politecn Nacl Ctr Invest & Estudios Avanzados Unidad Monterrey Biomol Div Lab Via Conocimiento 201 Apodaca 66600 Nuevo Leon Mexico Inst Tecnol Super Irapuato Ingn Sistemas Guanajuato Mexico Inst Politecn Nacl Ctr Invest & Estudios Avanzados Dept Computac Mexico City DF Mexico

Research on biology has seen significant advances with the use of molecular dynamics (MD) simulations. The MD methodology enables explanation and discovery of molecular mechanisms in a wide range of natural processes and biological systems. The need to readily share the ever-increasing amount of MD data has been hindered by the lack of specialized bioinformatic tools. The difficulty lies in the efficient management of the data, i.e., in sending and processing 3D information for its visualization. In this work, we present HTMoL, a plug-in-free, secure GPU-accelerated web application specifically designed to stream and visualize MD trajectory data on a web browser. Now, individual research labs can publish MD data on the Internet, or use HTMoL to profoundly improve scientific reports by including supplemental MD data in a journal publication. HTMoL can also be used as a visualization interface to access MD trajectories generated on a high-performance computer center directly. Furthermore, the HTMoL architecture can be leveraged with educational efforts to improve learning in the fields of biology, chemistry, and physics.

关键词： Data sharing Data publishing Data streaming graphical processing unit Webgl Web application Interactive structure Interactive dynamics

来源：评论

学校读者我要写书评

暂无评论

Time-dependent Density Functional-based Tight-bind Method Efficiently Implemented with OpenMP Parallel and GPU Acceleration

引用

Chinese Journal of Chemical Physics 2013年第6期26卷 635-645,I0003页

作者： Guo-hong Fan Ke-li Han Guo-zhong He State Key Laboratory of Molecular Reaction Dynamics Dalian Institute of Chemical Physics ChineseAcademy of Sciences Dalian 116023 China

The time-dependent density functional-based tight-bind （TD-DFTB） method is implemented on the multi-core and the graphical processing unit （GPU） system for excited state calcu-lations of large system with hundreds or thousands of atoms. Sparse matrix and OpenMP multithreaded are used for building the Hamiltonian matrix. The diagonal of the eigenvalue problem in the ground state is implemented on the GPUs with double precision. The GPU- based acceleration fully preserves all the properties, and a considerable total speedup of 8.73 can be achieved. A Krylov-space-based algorithm with the OpenMP parallel and CPU acceleration is used for finding the lowest eigenvalue and eigenvector of the large TDDFT matrix, which greatly reduces the iterations taken and the time spent on the excited states eigenvalue problem. The Krylov solver with the GPU acceleration of matrix-vector product can converge quickly to obtain the final result and a notable speed-up of 206 times can be observed for system size of 812 atoms. The calculations on serials of small and large systems show that the fast TD-DFTB code can obtain reasonable result with a much cheaper computational requirement compared with the first-principle results of CIS and full TDDFT calculation.

关键词： Density-functional theory Tight-binding method Time-dependent densityfunctional theory Excited state graphical processing unit Krylov iterative algorithm Sparse matrix OpenMP

来源：评论

学校读者我要写书评

暂无评论

TeraFLOP computing on a desktop PC with GPUs for 3D CFD

引用

INTERNATIONAL JOURNAL OF COMPUTATIONAL FLUID DYNAMICS 2008年第7期22卷 443-456页

作者： Toelke, J. Krafczyk, M. Tech Univ Carolo Wilhelmina Braunschweig Inst Comp Based Modeling Civil Engn Braunschweig Germany

A very efficient implementation of a lattice Boltzmann (LB) kernel in 3D on a graphical processing unit using the compute unified device architecture interface developed by nVIDIA is presented. By exploiting the explicit parallelism offered by the graphics hardware, we obtain an efficiency gain of up to two orders of magnitude with respect to the computational performance of a PC. A non-trivial example shows the performance of the LB implementation, which is based on a D3Q13 model that is described in detail.

关键词： lattice Boltzmann model D3Q13 model graphical processing unit high performance computing

来源：评论

学校读者我要写书评

暂无评论

Performance and energy-efficient implementation of a smart city application on FPGAs

引用

JOURNAL OF REAL-TIME IMAGE processing 2020年第3期17卷 729-743页

作者： Arif, Arslan Barrigon, Felipe A. Gregoretti, Francesco Iqbal, Javed Lavagno, Luciano Lazarescu, Mihai Teodor Ma, Liang Palomino, Manuel Segura, Javier L. L. Politecn Torino Turin Italy ACCIONA Infrastruct SA Madrid Spain

The continuous growth of modern cities and the request for better quality of life, coupled with the increased availability of computing resources, lead to an increased attention to smart city services. Smart cities promise to deliver a better life to their inhabitants while simultaneously reducing resource requirements and pollution. They are thus perceived as a key enabler to sustainable growth. Out of many other issues, one of the major concerns for most cities in the world is traffic, which leads to a huge waste of time and energy, and to increased pollution. To optimize traffic in cities, one of the first steps is to get accurate information in real time about the traffic flows in the city. This can be achieved through the application of automated video analytics to the video streams provided by a set of cameras distributed throughout the city. Image sequence processing can be performed both peripherally and centrally. In this paper, we argue that, since centralized processing has several advantages in terms of availability, maintainability and cost, it is a very promising strategy to enable effective traffic management even in large cities. However, the computational costs are enormous, and thus require an energy-efficient High-Performance Computing approach. Field Programmable Gate Arrays (FPGAs) provide comparable computational resources to CPUs and GPUs, yet require much lower amounts of energy per operation (around 6x for the application considered in this case study). They are thus preferred resources to reduce both energy supply and cooling costs in the huge datacenters that will be needed by Smart Cities. In this paper, we describe efficient implementations of high-performance algorithms that can process traffic camera image sequences to provide traffic flow information in real-time at a low energy and power cost.

关键词： Smart city Image processing Background subtraction Lucas-Kanade High-level synthesis Field programmable gate array graphical processing unit

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-Based Deep Learning Networks for Limited Data Wetland User Models 1

Uncertainty-Based Deep Learning Networks for Limited Data We...

引用

1st IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR)

作者： Hoblitzell, Andrew Babbar-Sebens, Meghna Mukhopadhyay, Snehasis Purdue Univ Dept Comp Sci W Lafayette IN 47907 USA Oregon State Univ Sch Civil & Construct Engn Corvallis OR 97331 USA IUPUI Dept Comp Sci Indianapolis IN 46202 USA

ISBN: (纸本)9781538692691

This paper discusses a method for dealing with limited data in deep networks based on calculating the uncertainty associated with remaining training data. The method was developed for the Watershed REstoration using Spatio-Temporal Optimization of REsources (WRESTORE) system, an interactive decision support system designed for performing multi-criteria decision analysis with a distributed system of conservation practices on the Eagle Creek Watershed in Indiana, USA. Our results show faster and more stable convergence when using an uncertainty-based incremental sampling method than when using a standard random incremental sampling method. This work describes the existing WRESTORE system, provides details about the implementation of our uncertainty-based incremental sampling method, and provides a discussion of our results and future work. The primary contribution of the paper is an uncertainty-based incremental sampling method which can be applied to limited data watershed design problems.

关键词： multiple criteria decision analysis limited data data science active learning optimal design incremental learning deep learning interactive machine learning design of experiments optimal design online learning optimal decisions systems engineering statistical process control adaptive learning iterative learning incremental learning uncertainty quantification simulation and modeling statistical efficiency Tensorflow massively parallel learning graphical processing unit

来源：评论

学校读者我要写书评

暂无评论

GPU-based Concurrent Multi-UAVs Path Flow Analysis using K-MST Algorithm 1

GPU-based Concurrent Multi-UAVs Path Flow Analysis using K-M...

引用

1st International Conference on Unmanned Vehicle Systems-Oman (UVS)

作者： Chauhan, Munesh Singh Coll Appl Sci Dept Informat Technol Salalah Oman

ISBN: (纸本)9781538693681

Unmanned Aerial Vehicles have widespread applications in the fields of remote sensing, reconnaissance, product delivery, and search & rescue. With the advent of new technologies the manufacturing cost of typical UAVs have gone down whereas on the contrary the robustness and safety factors have improved drastically. Hence multiple UAV systems are being planned for coordinating specific large tasks such as search & rescue and delivering commercial products door to door. Many factors affect the optimized operation of these multiple flying machines in a given designated geographical area. The most crucial being the challenge of chartering an optimal flight path for each UAV based on distance, terrain and fuel costs. Subsequently, cumulative flying time costs are calculated and a final scenario is presented highlighting the optimal usage of multiple UAVs working in unison. The prime focus is on finding localized shortest path using Prim's algorithm and then clustering these localized paths into smaller sub-graphs optimally. Multicore processors are leveraged to provide accelerated computations as the paths change dynamically with respect to time.

关键词： UAV graphical processing unit CUDA Prim's Algorithm K-MST clustering

来源：评论

学校读者我要写书评

暂无评论

Fast Computation Of Scalar Optical Diffraction Pattern For Pixelated Spatial Light Modulators

Fast Computation Of Scalar Optical Diffraction Pattern For P...

引用

22nd IEEE Signal processing and Communications Applications Conference (SIU)

作者： Esmer, Gokhan Bora Marmara Univ Elekt & Elekt Muhendisligi Bolumu Istanbul Turkey

ISBN: (纸本)9781479948741

Spatial light modulators can be used for having dynamic three dimensional images. Diffraction pattern of three dimensional objects can be used for driving those imaging devices. Diffraction patterns have to be calculated rapidly to form dynamic three dimensional images. Furthermore, pixelated structure of the spatial light modulators have to be taken into account in the diffraction pattern calculation to improve the quality of the reconstructed objects. The proposed method for calculating diffraction patterns not only takes into account the pixelated structure of the spatial light modulator to obtain more accurate results, but also provides fast calculation of the diffraction pattern. The proposed method employs a look-up table which is formed by 469 pre-computed vectors and the computation time of diffraction pattern can be decreased significantly compared to standard diffraction pattern calculation method. The proposed method can be run in parallel, hence further reduction in calculation time can be achieved and even the diffraction pattern of a three dimensional object which is formed by 3144 points can be calculated within 80msec when the algorithm is run in parallel on a graphics processing unit.

关键词： Computer-Generated Holography Spatial Light Modulator Look-Up Table Parallel processing graphical processing unit

来源：评论

学校读者我要写书评

暂无评论

Hunting the pertinency of hash and bloom filter combinations on GPU for fast pattern matching

引用

International Journal of Information Technology (Singapore) 2022年第5期14卷 2667-2679页

作者： Bhat, Radhakrishna Thilak, Reddy Kanala Vaibhav, Reddy Panyala Department of Computer Science and Engineering Manipal Institute of Technology Manipal Academy of Higher Education Karnataka Manipal 576104 India

There has been rapid growth in the field of graphical processing unit (GPU) programming due to the drastic increase in the computing hardware manufacturing. The technology used in these devices is now more affordable and accessible to the general public. With this growth, many serial programming applications that are now being transformed into more efficient parallel programming applications with significant improvement in the performance. The best example for this is parallel implementation of the probabilistic data structure Bloom filter in set membership queries. However, despite of it’s remarkable performance in speed and memory usage, there is a computational overhead in the calculation of hashes in Bloom filter. In this paper, the impact of the choice of hash functions on the qualitative properties of the Bloom filter has been experimentally recorded and the results show that there is a possibility of large performance gap among various hash functions. We have implemented the Bloom filter based pattern matching technique on GPU using compute unified device architecture (CUDA) and benchmark the performance of several cryptographic and non-cryptographic hash functions. © 2022, The Author(s).

关键词： Bloom filter graphical processing unit Parallel computing Parallel programming Pattern matching

来源：评论

学校读者我要写书评

暂无评论

POTENTIAL FOR INTERACTIVE DESIGN SIMULATIONS IN DISCRETE ELEMENT MODELLING 5

POTENTIAL FOR INTERACTIVE DESIGN SIMULATIONS IN DISCRETE ELE...

引用

5th International Conference on Particle-Based Methods - Fundamentals and Applications (PARTICLES)

作者： Wilke, Daniel N. Govender, Nicolin Pizette, Patrick Rajamani, Raj K. Univ Pretoria Ctr Asset & Integr Management ZA-0086 Pretoria South Africa Res Ctr Pharmaceut Engn RCPE GmbH Graz Res Ctr Pharmaceut Engn Graz Austria CSIR CHPC Cape Town South Africa Univ Lille EA 4515 IMT Lille Douai LGCgE Lab Genie Civil & geoEnvironmDept Genie Ci F-59000 Lille France Univ Utah Dept Met Engn Salt Lake City UT 84112 USA

ISBN: (纸本)9788494690976

This study investigates the potential for combining lower fidelity models with high performance solution strategies such as efficient graphical processing unit (GPU) based discrete element modelling (DEM) to not only do simulations faster but differently. Specifically this study investigates interactive simulation and design for which the simulation environment B1azeDEM-GPU was developed that allows researchers and engineers to interact with simulations. The initial results prove to be promising and warranting extensive research to be conducted in future which may allow for the development of alternative paradigms. In addition to the design cycle, the role that this interactive simulation and design will play in education is invaluable as an in-house corporate training tool for young engineers to actively train and develop understanding for specific industrial processes. This would also allow engineers to conduct just-in-time (JIT) simulation based assessment of processes before commencing on actual site visits, allowing for shorter and more focussed site excursions.

关键词： DEM graphical processing unit Interactive Design Interactive Simulation Tumbling Mill Numerical Multi-Fidelity Models

来源：评论

学校读者我要写书评

暂无评论

3D Object Matching on the GPU Using Spin-Image Surface Matching

3D Object Matching on the GPU Using Spin-Image Surface Match...

引用

Conference on Advanced Signal processing Algorithms, Architectures, and Implementantions XVIII

作者： Davis, Nolan Braunreiter, Dennis Tebcherani, Cezario Tanida, Masatoshi Sci Applicat Int Corp San Diego CA 92121 USA

ISBN: (纸本)9780819472946

Spin-image surface matching is a technique for locating objects in a scene by processing three-dimensional surface information from sources such as light detection and ranging (LIDAR), structured light photography, and tomography. It is attractive for parallel processing on graphics processing units (GPUs) because the two main computational steps matching pairs of spin-images by correlation, and matching pairs of points between model and scene - are explicitly parallel. By implementing these parallel computations on the GPU, as well as recasting serial portions of the algorithm into a parallel form and structuring the algorithm to limit data exchanges between host and GPU, this project achieved an overall speedup of 20 times or more compared to conventional serial processing. A demonstration application has been developed that allows users to select among a set of models and scenes and then applies the spin-image surface matching algorithm to match the selected models to the scene. It also has several user interface controls for changing parameters. One new parameter is a geometric consistency ratio (GCR) that quantifies the matching performance and provides a measure for discarding low-quality matches. By toggling between GPU- and host-based processing, the application demonstrates the speedup achieved with parallelization on the GPU.

关键词： spin image surface matching object matching image correlation geometric consistency GPU graphical processing unit parallel processing speedup

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：