检索结果-内蒙古大学图书馆

Dynamic multi-level load balancing for scalable simulations of reacting multiphase flows

international JOURNAL OF high performance computing APPLICATIONS 2025年

作者： van den Oord, Gijs Azizi, Victor Fathi, Mohamad Hickel, Stefan Netherlands Esci Ctr Nat Sci & Engn Sect Sci Pk 402 NL-1098 XH Amsterdam Netherlands Delft Univ Technol Fac Aerosp Engn Aerodynam Grp Delft Netherlands

simulations of reacting multiphase flows tend to display an inhomogeneously distributed computational intensity over the spatial and temporal domains. The time-to-solution of chemical reaction rates can span multiple orders of magnitude due to the emergence of combustible kernels and thin turbulent reaction zones. Similarly, the time to solve the equation of state (EoS) for non-ideal fluid mixtures deviates substantially between the grid cells. These effects result in a performance profile that is unbalanced and rapidly changing for transient simulations, and therefore beyond the capabilities of traditional (quasi-)static mesh partitioning methods. We analyse this loss of parallel efficiency for large-eddy simulations of the ECN Spray-A benchmark with the multi-physics solver INCA and propose to mitigate the problem by introducing two independent repartitioning stages in addition to the classic domain decomposition for fluid transport: one for the EoS and one for chemical reactions. We explore various scalable repartitioning strategies in this context and observe that rebalancing computational load yields a significant speedup that is robust for various mesh resolutions and process numbers. The dynamic multistage load-balancing thus effectively removes obstacles towards good parallel scaling of INCA and similar solvers for reacting and/or multiphase flows.

关键词： Load balancing parallel computing large-eddy simulation reacting flow multiphase flow ECN spray-A

来源：评论

学校读者我要写书评

暂无评论

Quantifying performance Trade-offs in Network Virtualization for Cloud computing Environments

Quantifying Performance Trade-offs in Network Virtualization...

引用

2025 international conference on Pervasive Computational Technologies, ICPCT 2025

作者： Venkatesha, Guruprasad Govindappa Sharma, Rashmi Marella, Venkatadri Acharya, Vinay Sivaraj, Krishna Prasath Shah, Samarth Bms College of Engineering Basavanagudi Karnataka Bengaluru India Symbiosis Institute of Technology Nagpur Campus Pune India International Technological University Santa ClaraCA United States University of Texas at Dallas RichardsonTX United States The University of Toledo OH United States University at Albany AlbanyNY12222 United States

ISBN: (纸本)9798331508685

Designing a cloud computing environment is a tradeoff performance between costs and way of providing services that System of virtualized networks may affect it. In order to alleviate this, abstractions have to be developed to quantitively measure and decide data plane abstraction performance costs in the cloud computing spaces. Here we provide a technical summary of the work that was aimed at quantifying performance tradeoffs in cloud computing settings from the angle of network virtualization. This study will consider the performance of two such important factors like network latency and resource utilization. Network latency: This is defined as the time taken for a given network packet to travel from a given source to the given destination. Resource utilization, on the other hand, refers to the amount of network resource that used, e.g. how much bandwidth or computational power is used. This will then be supplemented by control the experiment to perform simulations for understanding network configuration effects and simulation threats. This will help in keeping a picture of how these trade-offs affect the overall performance in network virtualization. This study's findings will help network and cloud computing architects in optimizing their networks and alleviating the performance penalties of virtualization. Consequently, this helps in reducing costs and enhancing performance in cloud computing environments. © 2025 IEEE.

关键词： Virtual environments

来源：评论

学校读者我要写书评

暂无评论

Water distribution network calibration for unreported leak localization with consideration of uncertainties

引用

international JOURNAL OF ENVIRONMENTAL SCIENCE AND TECHNOLOGY 2025年第1期22卷 399-418页

作者： Moasheri, R. Ghazizadeh, M. Jalili Kohanali, R. Ahmadi Shahid Beheshti Univ Fac Civil Water & Environm Engn Tehran Iran

Leakage in water distribution networks precipitates both water wastage and the ingress of pollutants. The localization of leaks, a formidable challenge within water demand management, has spurred an examination of hydraulic simulation-based methodologies as a more economically feasible and time-efficient alternative to conventional methods. This paper introduces a framework for precisely determining the location of leaks within a water distribution network, leveraging the Grasshopper Optimization Algorithm. The approach meticulously compares simulated data with pressure field information. Acknowledging the intrinsic uncertainties pertaining to hydraulic model parameters-such as elevations, nodal base demand, and pipe roughness coefficients in real-world water distribution networks-the developed method incorporates perturbation analysis for judicious parameter selection. Monte Carlo simulation is then employed to apply these parameters in the simulation process systematically. The efficacy of the method is demonstrated by applying it to benchmark water distribution networks (specifically, Poulakis and Balerma) under various leakage scenarios, achieving accuracy levels of up to 99%. Introducing uncertainty into the simulation process results in a maximum 20% reduction in method accuracy. Real-world implementation successfully and accurately localizes leakage, affirming the practical applicability of the proposed method for water utilities.

关键词： Grasshopper optimization algorithm K-means clustering Leakage detection Monte Carlo simulation Water distribution networks Water loss

来源：评论

学校读者我要写书评

暂无评论

Innovations in mathematical modeling, AI, and optimization techniques

引用

JOURNAL OF SUPERcomputing 2025年第1期81卷 1-4页

作者： Ohue, Masahito Yasuo, Nobuaki Takata, Masami Inst Sci Tokyo Sch Comp Dept Comp Sci Yokohama Kanagawa 2268501 Japan Inst Sci Tokyo Acad Convergence Mat & Informat TAC MI Tokyo 1528550 Japan Nara Womens Univ Res Grp Informat & Commun Technol Life Nara 6308506 Japan

This special issue is dedicated to examining the rapidly evolving fields of artificial intelligence, mathematical modeling, and optimization, with particular emphasis on their growing importance in computational science. It features the most notable papers from the "Mathematical Modeling and Problem Solving" workshop at PDPTA'24, the 30th international conference on Parallel and Distributed Processing Techniques and Applications. The issue showcases pioneering research in areas such as natural language processing, system optimization, and high-performance computing. The nine selected studies include novel AI-driven methods for chemical compound generation, historical text recognition, and music recommendation, along with advancements in hardware optimization through reconfigurable accelerators and vector register sharing. Additionally, evolutionary and hyper-heuristic algorithms are explored for sophisticated problem-solving in engineering design, and innovative techniques are introduced for high-speed numerical methods in large-scale systems. Collectively, these contributions demonstrate the significance of AI, supercomputing, and advanced algorithms in driving the next generation of scientific discovery.

关键词： Mathematical modeling Artificial intelligence Parallel and distributed computing Reconfigurable computing Drug discovery

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Parallelism Framework of SPH for the Applications in Automobile Gearbox 30th

A Hybrid Parallelism Framework of SPH for the Applications i...

引用

30th international conference on Computational and Experimental Engineering and Sciences

作者： Zhang, Xiang Sun, Peng-Nan Xu, Yang Sun Yat Sen Univ Sch Ocean Engn & Technol Zhuhai 519082 Peoples R China

ISBN: (纸本)9783031774911;9783031774898;9783031774881

The rotating machineries of automobile, such as gears and motors, are involved with complicated interactions between fluids and structures, resulting in flow phenomenon such as free-surface, moving boundary, thermal conduction etc. Smoothed Particle Hydrodynamics (SPH), due to its Lagrangian nature, is preferred to simulate such phenomenon. The complexities of automobile structures require small particle distancing and therefore large number of particles are needed to discretize both fluid and structure. The small time step of SPH simulation is also required due to intense flowsplashing resulting from high-speedmoving boundary. Both two points above lead to large amount of computation duringSPHsimulation. In this paper, a parallelism framework of weakly compressible SPH(WCSPH) is proposed to accelerate SPH simulation by high-performance computing cluster. A hybrid parallelism strategy, with both Message Passing Interface (MPI) and Intel Threading Building Blocks (TBB), is used to reduce the total number of processes and therefore reduce the latency due to communication among computing clusters network. METIS is used to decompose the computational domain enabling dynamic domain decomposition and load balancing. The oil motion inside a gearbox is successfully simulated using the proposed framework, showing that the proposed parallelism framework is applicable to complex industry application and can accelerate SPH simulation efficiently.

关键词： SPH parallel computing fluid-structure interaction gearbox

来源：评论

学校读者我要写书评

暂无评论

Exploring Hidden Behaviors in OpenMP Multi-threaded Applications for Anomaly Detection in HPC Environments 21st

Exploring Hidden Behaviors in OpenMP Multi-threaded Applicat...

引用

21st international conference on Distributed computing and Intelligent Technology

作者： Bhowmik, Biswajit Girish, K. K. Mishra, Pawanesh Mishra, Rishi Natl Inst Technol Dept Comp Sci & Engn Maharshi Sushrut CAS Lab BRICS Lab Mangalore Karnataka India

ISBN: (纸本)9783031814037;9783031814044

In high-performance computing (HPC), multi-threaded applications using OpenMP face complex challenges in identifying hidden performance issues, often due to resource conflicts, software inefficiencies, and hardware anomalies. These subtle issues can significantly degrade performance and reduce system reliability. This paper introduces an innovative approach designed to address these concealed issues in OpenMP multi-threaded applications. The proposed method integrates a Random Forest classifier with anthropomorphic diagnosis to effectively identify and diagnose performance-affecting problems. The approach has demonstrated a remarkable ability to detect 90% of performance-affecting issues that are often obscured within complex HPC environments.

关键词： high performance computing (HPC) OpenMP Random Forest Anthropomorphic Diagnosis performance Issues System Efficiency

来源：评论

学校读者我要写书评

暂无评论

p-adaptive discontinuous Galerkin method for the shallow water equations on heterogeneous computing architectures

引用

GEM-international JOURNAL ON GEOMATHEMATICS 2025年第1期16卷 1-25页

作者： Faghih-Naini, Sara Aizinger, Vadym Kuckuk, Sebastian Angersbach, Richard Koestler, Harald Univ Bayreuth Chair Sci Comp D-95440 Bayreuth Germany Friedrich Alexander Univ Erlangen Nurnberg Chair Comp Sci 10 D-91058 Erlangen Germany Friedrich Alexander Univ Erlangen Nurnberg Erlangen Natl High Performance Comp Ctr NHRFAU D-91058 Erlangen Germany

Heterogeneous computing and exploiting integrated CPU-GPU architectures has become a clear current trend since the flattening of Moore's Law. In this work, we propose a numerical and algorithmic re-design of a p-adaptive quadrature-free discontinuous Galerkin (DG) method for the shallow water equations. Our new approach separates the computations of the non-adaptive (lower-order) and adaptive (higher-order) parts of the discretization from each other. Thereby, we can overlap computations of the lower-order and the higher-order DG solution components. Furthermore, we investigate execution times of main computational kernels and use automatic code generation to optimize their distribution between the CPU and GPU. Several setups, including a prototype of a tsunami simulation in a tide-driven flow scenario, are investigated, and the results show that significant performance improvements can be achieved in suitable setups.

关键词： p-adaptivity Heterogeneous architectures GPU computing System-on-a-chip (SoC) Discontinuous Galerkin method Quadrature-free integration Shallow water equations 76-10

来源：评论

学校读者我要写书评

暂无评论

Research on Efficient Processing Algorithm for Engineering Cost Data Using Cloud computing Platform

Research on Efficient Processing Algorithm for Engineering C...

引用

2025 Asia-Europe conference on Cybersecurity, Internet of Things and Soft computing, CITSC 2025

作者： Li, Dantong Dalian University of Finance and Economics Dalian China

ISBN: (纸本)9798331504205

This paper proposes a research scheme for efficient processing algorithm of engineering cost data based on cloud computing platform, aiming to improve data processing efficiency by utilizing the high-performance computing capability and distributed characteristics of cloud computing. The algorithm design takes distributed compression algorithm as the core, combines data partitioning, parallel computing and storage optimization technology to achieve efficient storage and fast access to large-scale cost data. In the system implementation, a parallel algorithm architecture that adapts to multi-node environment is designed. The algorithm performance is evaluated by building a simulation model. The experimental results show that the method proposed in this paper is superior to traditional algorithms in data compression rate, processing speed and resource utilization. When processing 1TB cost data, the compression rate reaches 85%, and the processing efficiency is improved by about 60%. Data analysis further verifies the stability and scalability of the algorithm, and provides theoretical basis and technical support for intelligent data processing in the field of engineering cost. This study opens up a new path for engineering cost data processing, and the proposed algorithm and platform architecture can be widely used in other complex data scenarios. © 2025 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Quantum transport simulation of α-GeTe ferroelectric semiconductor transistors

引用

JOURNAL OF MATERIALS CHEMISTRY C 2025年第2期13卷 568-577页

作者： Li, Qiang Yang, Zongmeng Yang, Xingyue Zhou, Wenjing Yang, Chen Sun, Xiaotian Fang, Shibo Lu, Jing Hubei Minzu Univ Dept Phys Enshi 445000 Peoples R China Peking Univ State Key Lab Mesoscop Phys Beijing 100871 Peoples R China Peking Univ Sch Phys Beijing 100871 Peoples R China Univ Michigan Dept Stat Ann Arbor MI 48109 USA Luoyang Normal Univ Coll Chem & Chem Engn Luoyang 471934 Peoples R China Luoyang Normal Univ Henan Key Lab Funct Oriented Porous Mat Luoyang 471934 Peoples R China Collaborat Innovat Ctr Quantum Matter Beijing 100871 Peoples R China Peking Univ Beijing Key Lab Magnetoelectr Mat & Devices BKL ME Beijing 100871 Peoples R China Peking Univ Yangtze Delta Inst Optoelect Nantong 226010 Peoples R China Peking Univ Key Lab Phys & Chem Nanodevices Beijing 100871 Peoples R China

Ferroelectric semiconductor transistor is a newly proposed device that uses ferroelectric semiconductors as channel materials for integrated memory and computation. Currently, the main challenge in advancing ferroelectric semiconductor transistors (FeS-FETs) is finding ferroelectric channel materials that balance high performance with industrial production feasibility. In this work, we predict the performance of alpha-GeTe, a quasi-two-dimensional ferroelectric semiconductor with excellent compatibility with Si-based substrates, as a FeS-FET by ab initio quantum transport simulation. When taking negative capacitance technology and underlap structure into account, we find that alpha-GeTe ferroelectric semiconductor transistors can meet the international technology roadmap for semiconductors for high-performance standards for industrial-grade chip logic operations with a 5-nm channel length, and achieve a ferroelectric switch ratio of 228 at zero gate voltage. The memory window (0.9 V) of the 5-nm gate-length monolayer alpha-GeTe FeS-FETs is three times larger than that (0.3 V) of the alpha-In2Se3 ferroelectric semiconductor transistor. Our work suggests that alpha-GeTe is a strong candidate for the future industrial fabrication of FeS-FETs.

关键词： Ferroelectric devices

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Regression-Based Prediction for Improving performance and Energy Consumption in HPC Platforms 11th

Machine Learning Regression-Based Prediction for Improving P...

引用

11th Latin American conference on high performance computing

作者： Coelho, Micaella Ocana, Kary Pereira, Andre Porto, Alexandre Cardoso, Douglas O. Lorenzon, Arthur Oliveira, Rui Navaux, Philippe O. A. Osthoff, Carla Natl Lab Sci Comp LNCC Rio De Janeiro Brazil Univ Minho Campus Gualtar Braga Portugal HASLab INESC TEC Campus Gualtar Braga Portugal Univ Porto Fac Arts & Human Ctr Linguist Porto Portugal Univ Fed Rio Grande do Sul Inst Informat Porto Alegre RS Brazil

ISBN: (纸本)9783031800832;9783031800849

high-performance computing is pivotal for processing large datasets and executing complex simulations, ensuring faster and more accurate results. Improving the performance of software and scientific workflows in such environments requires careful analysis of their computational behavior and energy consumption. Therefore, maximizing computational throughput in these environments, through adequate software configuration and resource allocation, is essential for improving performance. The work presented in this paper focuses on leveraging regression-based machine learning and decision trees to analyze and optimize resource allocation in high-performance computing environments based on application's performance and energy metrics. Applied to a bioinformatics case study, these models enable informed decision-making by selecting the appropriate computing resources to enhance the performance of a phylogenomics software. Our contribution is to better explore and understand the efficient resource management of supercomputers, namely Santos Dumont. We show that the predictions for application's execution time using the proposed method are accurate for various amounts of computing nodes, while energy consumption predictions are less precise. The application parameters most relevant for this work are identified and the relative importance of each application parameter to the accuracy of the prediction is analysed.

关键词： Machine learning high-performance computing Scientific applications Bioinformatics Resource management

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：