检索结果-内蒙古大学图书馆

A High-Performance and Cost-Efficient Interconnection Network for High-Density Servers

Journal of computer Science & Technology 2014年第2期29卷 281-292页

作者：包雯韬付斌章陈明宇张立新 State Key Laboratory of Computer Architecture Institute of Computing TechnologyChinese Academy of Sciences University of Chinese Academy of Sciences

The high-density server is featured as low power, low volume, and high computational density. With the rising use of high-density servers in data-intensive and large-scale web applications, it requires a high-performance and cost-efficient intra-server interconnection network. Most of state-of-the-art high-density servers adopt the fully-connected intra-server network to attain high network performance. Unfortunately, this solution costs too much due to the high degree of nodes. In this paper, we exploit the theoretically optimized Moore graph to interconnect the chips within a server. Accounting for the suitable size of applications, a 50-size Moore graph, called Hoffman-Singleton graph, is adopted. In practice, multiple chips should be integrated onto one processor board, which means that the original graph should be partitioned into homogeneous connected subgraphs. However, the existing partition scheme does not consider above problem and thus generates heterogeneous subgraphs. To address this problem, we propose two equivalent-partition schemes for the Hoffman-Singleton graph. In addition, a logic-based and minimal routing mechanism, which is both time and area efficient, is proposed. Finally, we compare the proposed network architecture with its counterparts, namely the fully-connected, Kautz and Torus networks. The results show that our proposed network can achieve competitive performance as fully-connected network and cost close to Torus.

关键词： high-density server interconnection network Moore graph Hoffman-Singleton graph equivalent partition

来源：评论

学校读者我要写书评

暂无评论

Optimizing Parallel S_n Sweeps on Unstructured Grids for Multi-Core Clusters

引用

Journal of computer Science & Technology 2013年第4期28卷 657-670页

作者：闫洁谭光明孙凝晖 State Key Laboratory of Computer Architecture Institute of Computing Technology Chinese Academy of Sciences University of Chinese Academy of Sciences

In particle transport simulations, radiation effects are olden described by the discrete ordinates （Sn） form of Boltzmann equation. In each ordinate direction, the solution is computed by sweeping the radiation flux across the grid. Parallel Sn sweep on an unstructured grid can be explicitly modeled as topological traversal through an equivalent directed acyclic graph （DAG）, which is a data-driven algorithm. Its traditional design using MPI model results in irregular communication of massive short messages which cannot be efficiently handled by MPI runtime. Meanwhile, in high-end HPC cluster systems, multicore has become the standard processor configuration of a single node. The traditional data-driven algorithm of Sn sweeps has not exploited potential advantages of multi-threading of multicore on shared memory. These advantages, however, as we shall demonstrate, could provide an elegant solution resolving problems in the previous MPI-only design. In this paper, we give a new design of data-driven parallel Sn sweeps using hybrid MPI and Pthread programming, namely Sweep-H, to exploit hierarchical parallelism of processes and threads. With special multi-threading techniques and vertex schedule policy, Sweep-H gets more efficient communication and better load balance. We further present an analytical performance model for Sweep-H to reveal why and when it is advantageous over former MPI counterpart. On a 64-node multicore cluster system with 12 cores per node, 768 cores in total, Sweep-H achieves nearly linear scalability for moderate problem sizes, and better absolute performance than the previous times speedup on 64 nodes）. MPI algorithm on more than 16 nodes （by up to two

关键词： parallel Sn sweep unstructured grid data-driven algorithm

来源：评论

学校读者我要写书评

暂无评论

Accelerating DNN-based 3D point cloud processing for mobile computing

引用

Science China(Information Sciences) 2019年第11期62卷 40-50页

作者： Bosheng LIU Xiaoming CHEN Yinhe HAN Jiajun LI Haobo XU Xiaowei LI State Key Laboratory of Computer Architecture Institute of Computing TechnologyChinese Academy of Sciences University of Chinese Academy of Sciences

3D point cloud data, which are produced by various 3 D sensors such as LIDAR and stereo cameras, have been widely deployed by industry leaders such as Google, Uber, Tesla, and Mobileye, for mobile robotic applications such as autonomous driving and humanoid robots. Point cloud data, which are composed of reliable depth information, can provide accurate location and shape characteristics for scene understanding, such as object recognition and semantic segmentation. However, deep neural networks(DNNs), which directly consume point cloud data, are particularly computation-intensive because they have to not only perform multiplication-and-accumulation(MAC) operations but also search neighbors from the irregular 3 D point cloud data. Such a task goes beyond the capabilities of general-purpose processors in realtime to figure out the solution as the scales of both point cloud data and DNNs increase from application to application. We present the first accelerator architecture that dynamically configures the hardware onthe-fly to match the computation of both neighbor point search and MAC computation for point-based DNNs. To facilitate the process of neighbor point search and reduce the computation costs, a grid-based algorithm is introduced to search neighbor points from a local region of grids. Evaluation results based on the scene recognition and segmentation tasks show that the proposed design harvests 16.4× higher performance and saves 99.95% of energy than an NVIDIA Tesla K40 GPU baseline in point cloud scene understanding applications.

关键词： deep neural network acceleration point cloud data neighbor point search mobile robotics hardware architecture

来源：评论

学校读者我要写书评

暂无评论

YOLO-LLTS: Real-Time Low-Light Traffic Sign Detection via Prior-Guided Enhancement and Multi-Branch Feature Interaction

arXiv

引用

arXiv 2025年

作者： Lin, Ziyu Wu, Yunfan Ma, Yuhang Chen, Junzhou Zhang, Ronghui Wu, Jiaming Yin, Guodong Lin, Liang Guangdong Key Laboratory of Intelligent Transportation System School of intelligent systems engineering Sun Yat-sen University Guangzhou510275 China Department of Architecture and Civil Engineering Chalmers University of Technology Sven Hultins gata 6 GothenburgSE-412 96 Sweden School of Mechanical Engineering Southeast University Nanjing211189 China School of Computer Science and Engineering Sun Yat-sen University Guangzhou510275 China

Detecting traffic signs effectively under low-light conditions remains a significant challenge. To address this issue, we propose YOLO-LLTS, an end-to-end real-time traffic sign detection algorithm specifically designed for low-light environments. Firstly, we introduce the High-Resolution Feature Map for Small Object Detection (HRFM-TOD) module to address indistinct small-object features in low-light scenarios. By leveraging high-resolution feature maps, HRFM-TOD effectively mitigates the feature dilution problem encountered in conventional PANet frameworks, thereby enhancing both detection accuracy and inference speed. Secondly, we develop the Multi-branch Feature Interaction Attention (MFIA) module, which facilitates deep feature interaction across multiple receptive fields in both channel and spatial dimensions, significantly improving the model’s information extraction capabilities. Finally, we propose the Prior-Guided Enhancement Module (PGFE) to tackle common image quality challenges in low-light environments, such as noise, low contrast, and blurriness. This module employs prior knowledge to enrich image details and enhance visibility, substantially boosting detection performance. To support this research, we construct a novel dataset, the Chinese Nighttime Traffic Sign Sample Set (CNTSSS), covering diverse nighttime scenarios, including urban, highway, and rural environments under varying weather conditions. Experimental evaluations demonstrate that YOLO-LLTS achieves state-of-the-art performance, outperforming the previous best methods by 2.7% mAP50 and 1.6% mAP50:95 on TT100K-night, 1.3% mAP50 and 1.9% mAP50:95 on CNTSSS, and achieving superior results on the CCTSDB2021 dataset. Moreover, deployment experiments on edge devices confirm the real-time applicability and effectiveness of our proposed approach. Copyright © 2025, The Authors. All rights reserved.

关键词： Traffic signs

来源：评论

学校读者我要写书评

暂无评论

Resource scheduling virtualization in service-oriented future Internet architecture

引用

The Journal of China Universities of Posts and Telecommunications 2015年第4期22卷 92-100页

作者： Li Lan Chen Jianya Cui Hongyan Huang Tao Liu Yunjie State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing Key Laboratory of Network System Architecture and Convergence Beijing University of Posts and Telecommunications

Service-oriented future internet architecture（SOFIA） is a clean-slate network architecture. In SOFIA, a service request is mainly processed through service resolution and network resource allocation. To realize the network resource allocation, we reference the idea of network virtualization and propose resource scheduling virtualization. In resource scheduling virtualization, a service request is abstracted as a virtual network（VN） and the network resources are allocated by mapping the VN onto the physical network. Resource scheduling virtualization provides centralized resource scheduling control within an autonomous system（AS） and achieves better controllability compared with the distributed schemes. Besides, resource scheduling virtualization supports multi-site selection as well. Meanwhile, we propose a collection of resource scheduling algorithms based on maximum resource tree（MRT） adapting to different scenarios. According to the simulation results, the proposed algorithms show good performance on the key metrics, such as acceptance ratio, revenue, cost and utilization. Moreover, the simulation results reveal that our algorithm is more efficient than the traditional ones.

关键词： SOFIA resource scheduling virtualization virtual network embedding maximum resource tree(MRT)

来源：评论

学校读者我要写书评

暂无评论

Interference of human's ankle to the channel characteristics of near surface WUSN

引用

The Journal of China Universities of Posts and Telecommunications 2014年第1期21卷 36-42页

作者： LI Li WU Wei CHEN Jian-ya LIU Yun-jie State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing Key Laboratory of Network System Architecture and Convergence Beijing University of Posts and Telecommunications

Because of its potential applications in agriculture, environment monitoring and so on, wireless underground sensor network（WUSN） has been researched more and more extensively in recent years. The main and most important difference of WUSN to terrestrial wireless sensor network（WSN） is the channel characteristics, which determines the design methodology of it. In this paper, the propagation character of electromagnetic（EM） wave in the near surface WUSN is analyzed, as well as the path loss model of it is given. In addition, the influence of human＇s ankle to the channel characteristics of near surface WUSN is investigated by electromagnetic theory analysis, simulation and experiment. A novel path loss model of near surface WUSN which takes the interference of human＇s ankle into consideration is proposed. It is verified that the existing of human above the WUSN system may cause additional attenuation to the signal of near surface WUSN which propagates as lateral wave along the ground. Moreover, the relation of the attenuation and operating frequency is deduced, which gives a reference to extend the frequency band applied in WUSN.

关键词： WUSN lateral wave human tissue path loss model EM wave attenuation

来源：评论

学校读者我要写书评

暂无评论

Temperature behaviors of transparent solar PV panels under various operating modes: an experimental and numerical study

引用

Renewable Energy 2025年 250卷

作者： Dong, Qichang Wu, Xuejiao Song, Ye Du, Yaohan Qi, Jiacheng Huang, Lin Li, Wengui Huang, Youbo Shi, Long State Key Laboratory of Fire Science University of Science and Technology of China Anhui Hefei230026 China GCL System Integration Technology Co Ltd. Jiangsu Suzhou China School of Architecture and Civil Engineering Chengdu University Sichuan Chengdu610106 China Centre for Infrastructure Engineering and Safety School of Civil and Environmental Engineering The University of New South Wales NSW2052 Australia College of Safety Engineering Chongqing University of Science and Technology Chongqing401331 China

Temperature is the key factor in determining the practical performance and energy production prediction of solar photovoltaic (PV) panels. Transparent solar PV panels, in contrast to traditional opaque solar PV panels, have grown rapidly. However, the temperature behaviors of transparent PV panels and the effects of operating modes (i.e., operating and non-operating) on their temperature behaviors remain unidentified. Therefore, this study investigated the temperature behaviors of transparent PV panels (i.e., hollow and non-hollow) and the effect of different operating modes on their thermal performance. The hollow layer is the primary component influencing the temperature behaviors of hollow and non-hollow PV panels. At a solar irradiance of 1000 W/m2, the hollow layer resulted in the steady-state temperature and the mean temperature rise rate on the exposed surface of hollow PV panels being 12.0 % and 5.6 % lower, respectively, compared to non-hollow PV panels. Hollow PV panels have better heat resistance under high solar irradiance conditions. Experimental results showed that the effects of the operating mode on the rate of change of the PV panel surface temperature can be ignored. A theoretical model was also developed for the first time, and its predictions obeyed well with the experimental data. These findings and the developed theoretical model offer a foundation for forecasting the temperature and energy generation rate of transparent solar PV panels, helping with future policymaking. © 2025 Elsevier Ltd

关键词： Energy generation Online or offline Solar irradiance Temperature characteristics Transparent solar PV panels

来源：评论

学校读者我要写书评

暂无评论

Steady state speed distribution analysis for a combined cellular automaton traffic model

引用

Chinese Physics B 2008年第8期17卷 2850-2858页

作者：王俊峰陈桂生刘进 College of Computer Science Sichuan University Institute of Electronic System Engineering State Key Laboratory of Software Engineering Wuhan University

Cellular Automaton （CA） based traffic flow models have been extensively studied due to their effectiveness and simplicity in recent years. This paper develops a discrete time Markov chain （DTMC） analytical framework for a Nagel-Schreckenberg and Fukui Ishibashi combined CA model （W^2H traffic flow model） from microscopic point of view to capture the macroscopic steady state speed distributions. The inter-vehicle spacing Maxkov chain and the steady state speed Markov chain are proved to be irreducible and ergodic. The theoretical speed probability distributions depending on the traffic density and stochastic delay probability are in good accordance with numerical simulations. The derived fundamental diagram of the average speed from theoretical speed distributions is equivalent to the results in the previous work.

关键词： cellular automaton traffic flow model speed distribution discrete time Markov chain

来源：评论

学校读者我要写书评

暂无评论

Physical Implementation of the 1GHz Godson-3 Quad-Core Microprocessor

引用

Journal of computer Science & Technology 2010年第2期25卷 192-199页

作者：范宝峡杨梁王江嵋王茹肖斌徐英刘动赵继业 Key Laboratory of Computer System and Architecture Institute of Computing TechnologyChinese Academy of Sciences Graduate University of Chinese Academy of Sciences Loongson Technology Corporation Limited

The Godson-3A microprocessor is a quad-core version of the scalable Godson-3 multi-core series. It is physically implemented based on the 65 nm CMOS process. This 174 mm2 chip consists of 425 million transistors. The maximum frequency is 1GHz with a maximum power consumption of 15 W. The main challenges of Godson-3A physical implementation include very large scale, high frequency requirement, sub-micron technology effects and aggressive time schedule. This paper describes the design methodology of the physical implementation of Godson-3A, with particular emphasis on design methods for high frequency, clock tree design, power management, and on-chip variation （OCV） issue.

关键词： physical implementation design methodology on-chip variation (OCV) low power clock tree

来源：评论

学校读者我要写书评

暂无评论

Resource Allocation for Multi-Modal Semantic Communication in UAV Collaborative Networks

引用

IEEE Transactions on Communications 2025年

作者： Hu, Han Zhu, Xingwu Zhou, Fuhui Wu, Wei Hu, Rose Qingyang Zhu, Hongbo Nanjing University of Posts and Telecommunications Jiangsu Key Laboratory of Wireless Communications Nanjing210003 China Nanjing University of Posts and Telecommunications Engineering Research Center of Health Service System Based on Ubiquitous Wireless Networks Ministry of Education Nanjing210003 China Southeast University School of Cyber Science and Engineering Nanjing211189 China Nanjing University of Aeronautics and Astronautics College of Artificial Intelligence Nanjing210000 China Xidian University State Key Laboratory of Integrated Services Networks Xian710071 China Nanjing University of Posts and Telecommunications College of Communication and Information Engineering China Utah State University Department of Electrical and Computer Engineering LoganUT210003 United States

Semantic communication is envisioned as a potential communication paradigm enabled by artificial intelligence and is promising to break the Shannon limit for future 6G networks. This paradigm benefits unmanned aerial vehicles (UAVs) to conserve communication resources and minimize latency by only transmitting task-relevant semantic information. However, resource allocation in the multiple collaborative UAV scenarios remains unexplored, particularly regarding multi-modal semantic communication. To tackle this challenge, this paper investigates a semantic-aware intelligent resource allocation method for multi-UAV-assisted semantic communication networks in the UAV image-sensing task-oriented scenario. A multi-modal semantic communication framework with multi-UAV relay collaboration is developed. At the semantic level, a novel quality of experience (QoE) and the transmission cost model are introduced, based on which a semantic-aware resource allocation problem is formulated, aiming to maximize QoE while minimizing the transmission cost by jointly optimizing the UAV trajectory, the spectrum bandwidth, the transmit power and the number of the transmitted semantic symbols. To deal with optimization challenges involving hybrid variables and coordination among UAVs, a multi-UAV hybrid decision-controlled deep reinforcement learning (DRL) scheme is proposed. Simulation results demonstrate the effectiveness of the proposed scheme compared with the benchmark schemes in achieving a good balance between the QoE and the transmission cost. © 1972-2012 IEEE.

关键词： Deep reinforcement learning

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：