检索结果-内蒙古大学图书馆

International Symposium on parallel and distributed Computing

作者： Salman Ahmed Shaikh Hiroyuki Kitagawa Akiyoshi Matono Artificial Intelligence Research Center National Institute of Advanced Industrial Science and Technology Tokyo Japan International Institute for Integrative Sleep Medicine University of Tsukuba Tsukuba Japan

ISBN: (数字)9798350369199

ISBN: (纸本)9798350369205

Moving objects’ trajectory data is becoming increasingly available with the omnipresent GPS-equipped devices, e.g., smart phones, smart watches, etc. Identification of similar trajectories is a fundamental requirement of many real-world applications, for instance, car pooling, road planning, epidemic contact tracing. There exist a number of studies focusing on trajectory similarity search, where given a query trajectory, similar trajectories are searched from a trajectory store. GPS devices generate trajectories as a continuous data stream, highlighting the critical need for real-time and continuous trajectory similarity search. Thus, given a trajectories stream ($S_{\Gamma}$) and a trajectory store (R), we propose a distributed, continuous and real-time similarity search between $S_{\Gamma}$ and R. The proposed method employs an index-based strategy to facilitate effective similarity search, with the added capability of execution in distributed environments to ensure scalable processing. A comprehensive empirical study is provided to demonstrate the efficiency and pruning capability of the proposed approach.

关键词： Epidemics Roads Focusing real-time systems Trajectory Planning Object recognition distributed computing Smart phones Global Positioning System

来源：评论

学校读者我要写书评

暂无评论

Practical Tie-Breaking for parallel/distributed Simulations

Practical Tie-Breaking for Parallel/Distributed Simulations

引用

International workshop/Symposium on distributed Simulation and real-time Applications (DS-RT)

作者： Andrea Piccione Alessandro Pellegrini Huawei Munich Research Center Tor Vergata University of Rome

In this paper, we discuss a tie-breaking strategy based on a bitwise comparison of event payload that allows parallel and distributed discrete-event simulations to observe a deterministic order in the execution of events, even in the presence of event ties. This approach provides practical usability whenever model-assisted tie-breaking is unavailable, thus ensuring that multiple simulation executions provide deterministic behaviour and repeatable results. Moreover, it ensures that the selected order of events is also consistent with sequential executions. We discuss the theory behind this strategy and experimentally show that the performance drop is imputable to event queue management when relying on tie-breaking strategies like the ones discussed in this work.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Study of a Self-Tuning Predictive Voltage and Frequency Controller on a Cyber-Physical Model of a distributed Generation Plant

Study of a Self-Tuning Predictive Voltage and Frequency Cont...

引用

2023 Belarusian-Ural-Siberian Smart Energy Conference, BUSSEC 2023

作者： Bulatov, Yu.N. Kryukov, A.V. Suslov, K.V. Bratsk State University Bratsk Russia Irkutsk State Transport University Irkutsk National Research Technical University Irkutsk Russia National Research University 'Moscow Power Engineering Institute' Moscow Russia Transbaikal State University Chita Russia

ISBN: (纸本)9798350358070

The development of digital technology for monitoring and controlling physical processes of individual distributed generation plants in the electric power systems and its integration are driving the need to consider cyber-physical systems. Changes in operating conditions of distributed generation plants and in topology of electrical network require tuning the controllers to ensure stability of parallel operation of synchronous generators used. Controllers with a predictive device can be implemented to improve the efficiency of the tuning procedures due to their significant complexity. The paper provides a structure and a description of a digital model of a self-tuning predictive voltage and frequency controller for a synchronous generator. The studies were carried out on a cyber-physical model of a turbine generator set based on the MATLAB/Simulink real time Windows Target software. The purpose of the presented studies is to determine the effectiveness of the proposed self-tuning predictive controller in various transient conditions of the cyber-physical model of a turbine generator set in comparison with typical proportional- integral-derivative (PID) controllers. The experimental results demonstrated the advantages and effectiveness of the proposed predictive controller. In both normal and emergency conditions, it improved the quality indicators of the voltage and frequency control process. © 2023 IEEE.

关键词： Cyber Physical System

来源：评论

学校读者我要写书评

暂无评论

DTS: A Simulator to Estimate the Training time of distributed Deep Neural Networks 30

DTS: A Simulator to Estimate the Training Time of Distribute...

引用

30th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication systems (MASCOTS)

作者： Robinson, Wilfredo J. M. Esposito, Flavio Zuluaga, Maria A. St Louis Univ Dept Comp Sci St Louis MO 63103 USA Eurecom Data Sci Dept Biot France

ISBN: (纸本)9781665455800

Deep Neural Networks (DNNs) process big datasets achieving high accuracy on incredibly complex tasks. However, this progress has led to a scalability impasse, as DNNs require massive amounts of processing power and local memory to be trained, making them impossible or impractical to be used on a single device. This situation has led to the design of distributed training architectures, where the DNN and the training data can be split among multiple processors. How to choose the appropriate distributed training architecture, however, remains an open question. To help bring insights into this debate, in this work we design a distributed Training Simulator (DTS) that estimates the training time of a DNN in a distributed architecture through a mathematical model of the distributed architecture and resource-allocation heuristics. We illustrate the power of the proposed DTS through the implementation of five different distributed architectures, Pipeline Learning, Federated Learning, Split Learning, parallel Split Learning, and Federated Split Learning, and we validate the accuracy of the training estimates using three different datasets of varying complexity and two different DNNs. Finally, we present a trade-off analysis to demonstrate the coherence of DTS estimates for diverse high-performance computing scenarios by comparing these estimates with the behaviors of a real computer cluster.

关键词： big data distributed computing deep learning simulator

来源：评论

学校读者我要写书评

暂无评论

GraphA: Efficient Partitioning and Storage for distributed Graph Computation

引用

IEEE TRANSACTIONS ON SERVICES COMPUTING 2021年第1期14卷 155-166页

作者： Zhang, Yiming Li, Dongsheng Zhang, Chengfei Wang, Jinyan Liu, Ling Natl Univ Def Technol Sch Comp Changsha 410073 Peoples R China Georgia Inst Technol Coll Comp Atlanta GA 30332 USA

distributed graph computation is central to applications ranging from language processing to social networks. However, natural graphs tend to have skewed power-law distributions where a small subset of the vertices have a large number of neighbors. Existing graph-parallel systems suffer from load imbalance, high communication cost, and inefficient processing. To address this problem, in this paper we present GraphA, an adaptive scheme for efficient large-scale graph computation. At the core of GraphA is an adaptive and uniform graph partitioning algorithm, which partitions the datasets by using an incremental number of mapping functions. GraphA further improves and leverages the ART index structure to realize fine-grained and low-cost graph storage. Wehave implemented GraphA both on Spark and on GraphLab. Extensive evaluation shows that GraphA significantly outperforms state-of-the-art graph-parallel systems (GraphX and PowerLyra) in ingress time, execution time and storage cost, for both real-world and synthetic graphs. GraphA achieves up to 7.1x performance improvement over GraphX and 19.7 percent improvement over PowerLyra.

关键词： Natural graphs adaptive partitioning power-law distribution ART-indexed storage graph-parallel systems ingress time execution time

来源：评论

学校读者我要写书评

暂无评论

GCNScheduler: Scheduling distributed Computing Applications using Graph Convolutional Networks 1

GCNScheduler: Scheduling Distributed Computing Applications ...

引用

1st International workshop on Graph Neural Networking (GNNet)

作者： Kiamari, Mehrdad Krishnamachari, Bhaskar Univ Southern Calif Los Angeles CA 90007 USA

ISBN: (纸本)9781450399333

We provide a highly-efficient solution to the classical problem of scheduling task graphs corresponding to complex applications on distributed computing systems. A number of heuristics have been previously proposed to optimize task scheduling with respect to different metrics (e.g. makespan and throughput). However, they tend to be slow to run, particularly for larger problem instances, limiting their applicability in more dynamic systems. Motivated by the goal of solving these problems more rapidly, we propose, for the first time, a graph convolutional network-based scheduler (GCNScheduler). By carefully integrating the inter-task data dependency structure and the computational network into a single input graph, the GCNScheduler can efficiently schedule tasks of complex applications for a given objective. We use simulations to illustrate that not only can our scheme quickly and efficiently learn from existing scheduling schemes, but also it can easily be applied to large-scale settings that current scheduling schemes fail to handle. We demonstrate the generalization of GCNScheduler to unseen real-world applications and show that it achieves almost the same makespan and throughput as benchmarks, while providing several orders of magnitude faster scheduling times.

关键词： Graph Convolutional Networks GCNScheduler

来源：评论

学校读者我要写书评

暂无评论

Design, characterization, prototyping and digital twin of a Stewart platform. 16

Design, characterization, prototyping and digital twin of a ...

引用

16th Congress on Technology, Learning, and Teaching of Electronics (TAEE)

作者： Olmo, Alba Correal Gomez, Francisco J. Rios Univ Malaga Elect Robot & Mechatron Malaga Spain Univ Malaga Elect Dept Malaga Spain

ISBN: (纸本)9798350348682;9798350348675

This article aims to show the results of a Final Degree Project where electronics combines with other fields for the development of precision positioning systems. Thus, electronics, mechanics, computation and control come together in real time for innovation in these cutting-edge devices, whose development is a revolution for today's industry. This transversal methodology as the axis of electronics engineering is in line with the philosophy of the development of practices for students within the Department of Electronics of University of Malaga.

关键词： distributed Control Electronics Non Linear and parallel mechanics Digital Twin Stewart Platform Mechatronics

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Voltage/Current Control Scheme With Low-Communication Burden for Grid-Connected Series-Type Inverters in Decentralized Manner

引用

IEEE TRANSACTIONS ON POWER ELECTRONICS 2022年第1期37卷 920-931页

作者： Hou, Xiaochao Sun, Kai Zhang, Xin Sun, Yao Lu, Jinghang Tsinghua Univ Dept Elect Engn State Key Lab Power Syst Beijing 100084 Peoples R China Zhejiang Univ Coll Elect Engn Hangzhou 310027 Peoples R China Cent South Univ Sch Automat Hunan Prov Key Lab Power Elect Equipment & Grid Changsha 410083 Peoples R China Harbin Inst Technol Shenzhen Shenzhen 518055 Peoples R China

In this article, a novel hybrid voltage/current control scheme with low-communication burden is proposed for series-type inverters in a decentralized manner. All the inverter units are controlled by their individual local controllers without a real-time central global-communication. Among them, a few inverters serve as the current controlled inverters (CCIs) to regulate the grid current, and the rest inverters work as the voltage controlled inverters (VCIs). Under this control architecture, five main control targets are achieved: 1) frequency self-synchronization of each VCI inverter based on local current phase-angle signals;2) fault-tolerant capability of all inverters without one-to-all-failure;3) high disturbance-rejection-capability to grid variations;4) adjustable grid power factor;and 5) independent power control for inverters with unbalanced power sources. As the central high-bandwidth communication is avoided, it has the features of decreased cost, increased scalability, and reduced geographic restrictions of inverters. Compared with conventional decentralized approaches, only a few CCI units acquire the grid phase-angle signal to achieve grid synchronization by communication, and the most VCI units are local self-synchronization via the local line-current-phase signal rather than real-time grid-voltage-phase communication signal. Thus, the proposed control can be realized with a very low-communication burden, which is more cost-effective and higher reliability in terms of communication faults. The feasibility of the proposed method is verified by experimental results of eight different cases, such as source power change, large power gap, grid voltage/frequency deviation, grid harmonics, and one-fault redundancy.

关键词： Inverters Voltage control Synchronization Hybrid power systems Reactive power Regulation real-time systems Decentralized control distributed generation (DG) microgrid parallel inverters series inverters

来源：评论

学校读者我要写书评

暂无评论

Resource-Aware distributed Differential Evolution for Training Expensive Neural-Network-Based Controller in Power Electronic Circuit

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING systems 2022年第11期33卷 6286-6296页

作者： Liu, Xiao-Fang Zhan, Zhi-Hui Zhang, Jun Nankai Univ Inst Robot & Automat Informat Syst Coll Artificial Intelligence Tianjin 300350 Peoples R China Nankai Univ Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Peoples R China Pazhou Lab Guangzhou 510330 Peoples R China South China Univ Technol Guangdong Prov Key Lab Computat Intelligence & Cy Guangzhou 510006 Peoples R China Zhejiang Normal Univ Jinhua 321004 Zhejiang Peoples R China Chaoyang Univ Technol Taichung 41349 Taiwan

The neural-network (NN)-based control method is a new emerging promising technique for controller design in a power electronic circuit (PEC). However, the optimization of NN-based controllers (NNCs) has significant challenges in two aspects. The first challenge is that the search space of the NNC optimization problem is such complex that the global optimization ability of the existing algorithms still needs to be improved. The second challenge is that the training process of the NNC parameters is very computationally expensive and requires a long execution time. Thus, in this article, we develop a powerful evolutionary computation-based algorithm to find a high-quality solution and reduce computational time. First, the differential evolution (DE) algorithm is adopted because it is a powerful global optimizer in solving a complex optimization problem. This can help to overcome the premature convergence in local optima to train the NNC parameters well. Second, to reduce the computational time, the DE is extended to distribute DE (DDE) by dispatching all the individuals to different distributed computing resources for parallel computing. Moreover, a resource-aware strategy (RAS) is designed to further efficiently utilize the resources by adaptively dispatching individuals to resources according to the real-time performance of the resources, which can simultaneously concern the computing ability and load state of each resource. Experimental results show that, compared with some other typical evolutionary algorithms, the proposed algorithm can get significantly better solutions within a shorter computational time.

关键词： Artificial neural networks Optimization Training Power electronics Approximation algorithms Voltage measurement Steady-state distributed differential evolution (DDE) neural-network-based controller (NNC) power electronic circuit (PEC) resource-aware strategy (RAS)

来源：评论

学校读者我要写书评

暂无评论

Reconfigurable Event-Driven Spiking Neuromorphic Computing near High-Bandwidth Memory

Reconfigurable Event-Driven Spiking Neuromorphic Computing n...

引用

2023 IEEE Biomedical Circuits and systems Conference, BioCAS 2023

作者： Hota, Gopabandhu Frank, Gwenevere Wang, Keli Uppal, Abhinav Olajide, Omowuyi Liu, Jeffrey Bansal, Shashank Yoshimoto, Kenneth Wang, Qingbo Deiss, Stephen Cauwenberghs, Gert Uc San Diego Institute for Neural Computation La Jolla CA92093 United States

ISBN: (纸本)9798350300260

This paper presents the hardware architecture and processing dataflow of a general-purpose reconfigurable, spike event-driven computing platform for running bio-inspired neural networks in real-time. Near-memory computing architecture and hardware-aware compiler optimization facilitate low-latency, distributed parallel execution of user-specified network models of arbitrary connection topology. The massively parallel processing architecture scales to implementation of very large networks through hierarchical address-event routing (HiAER) of spikes implementing long-range synaptic connectivity across the network. Large-size bio-inspired AI model capacity and high-performance neuronal network emulation in parallel hardware are enabled by virtue of memory-efficient network storage and execution dataflow optimized for sparse networks. Implemented on a high-end Field Programmable Gate Array (FPGA), the HiAER-Spike platform captures axonal and neuronal events in internal FPGA memories which, when activated, fetch synaptic connectivity from lookup tables stored in High Bandwidth Memory (HBM) 2.0 to update the neuron membrane state variables. Threshold crossing events trigger output spikes which in turn propagate as axonal activity through the network. The system streams input and output spike events over a PCIe 3.0 interface with the host station. A compiler toolchain generates the necessary FPGA and HBM hardware configuration for partitioning and mapping of the network topology and synaptic connectivity onto the HBM data structure. The memory organization and microarchitecture of the FPGA implementation supports 4M axons, 4M neurons, and 1B synapses, and leverages the nominal 450GBps throughput of HBM 2.0 for a throughput of 0.1 tera synaptic operations per second (TSynOps). The architecture efficiently handles both sparse connectivity and sparse activity for robust and low-latency event-driven inference for both edge and cloud computing, with examples demonstrating spike

关键词： distributed computing neuromorphic systems engineering reconfigurable hardware Spiking neural networks

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：