检索结果-内蒙古大学图书馆

IEEE International Conference on Big Data

作者： Hooyoung Ahn Seonyoung Kim Yoomi Park Woojong Han Shinyoung Ahn Tu Tran Bharath Ramesh Hari Subramoni Dhabaleswar K. Panda Supercomputing System Research Section Electronics and Telecommunications Research Institute Department of Computer Science and Engineering The Ohio State University

ISBN: (数字)9798350362480

ISBN: (纸本)9798350362497

In Artificial Intelligence (AI) and high-performance computing (HPC), growing data and model sizes require distributed processing across multiple nodes due to single-node limitations, increasing inter-node communication. To address these challenges, we propose a novel MPI allgather method leveraging CXL technology, which supports composable architectures and dynamic resource allocation in data centers and HPC systems. Notably, CXL 3.1 facilitates cache coherence among nodes. The proposed allgather method uses the CXL shared memory pool as a communication buffer, outperforming existing algorithms for two reasons: First, CXL provides lower latency than Ethernet and IB, and second, by using the CXL shared memory pool as a shared communication buffer across multiple nodes, it significantly reduces the number of communications. To the best of our knowledge, this work is the first to explore combining MPI collective communication with CXL technology to optimize MPI allgather. Our proposed allgather method significantly reduces communication latency compared to traditional allgather methods by up to 42.14x, with a minimum improvement of 2.91x, as measured using the OSU Micro-Benchmark (OMB), a standard MPI benchmarking suite.

关键词： High performance computing Heuristic algorithms Memory management Ethernet Dynamic scheduling Hardware Topology Resource management Artificial intelligence Standards

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Load Balancing Algorithm for Stable Pushing of Docker Images in Nginx 6

A Dynamic Load Balancing Algorithm for Stable Pushing of Doc...

引用

6th International Conference on Machine Learning, Big Data and Business Intelligence, MLBDBI 2024

作者： Xi, Hajie Hu, Song Zhang, He Zhang, Shuang Weng, Jingnong Huang, Jian School of Computer Science and Engineering Beihang University Beijing China Research Institute of Highway Ministry of Transport State Key Lab of Transportation System Beijing China Transportation Comprehensive Enforcement Corps Beijing China School of Software Beihang University Beijing China

ISBN: (纸本)9798331541798

With the rapid development of artificial intelligence, a wide variety of deep learning models have emerged across various fields. Nevertheless, the quality of these models varies significantly, making the effective evaluation of model performance a pressing issue. In some open data competitions, remote evaluation based on Docker images has increasingly become a novel and reliable approach, providing a consistent environment to ensure the fairness of model evaluations. However, it imposes high demands on hardware resources, typically requiring dedicated resources for each participant to serve as Docker Registries for the pushing and management of Docker images. In this paper, we extend the load balancing module of the Nginx server by integrating Lua scripts and propose a Cache-based Dynamic Smooth Weighted Round-robin (CDSWRR) algorithm to optimize the pushing process of Docker images. By constructing a unified load balancing entry, the proposed approach simplifies the deployment of Docker Registry clusters. Experiments show that compared to weighted round-robin and IP hash algorithms of Nginx, the CDSWRR algorithm is able to ensure stable pushing of Docker images and reduce both response error rate and response time. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Jin, Dongyang Fan, Chao Chen, Weihua Yu, Shiqi Department of Computer Science and Engineering Southern University of Science and Technology China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Alibaba Group China

ISBN: (纸本)157735897X

The gait, as a kind of soft biometric characteristic, can reflect the distinct walking patterns of individuals at a distance, exhibiting a promising technique for unrestrained human identification. With largely excluding gait-unrelated cues hidden in RGB videos, the silhouette and skeleton, though visually compact, have acted as two of the most prevailing gait modalities for a long time. Recently, several attempts have been made to introduce more informative data forms like human parsing and optical flow images to capture gait characteristics, along with multi-branch architectures. However, due to the inconsistency within model designs and experiment settings, we argue that a comprehensive and fair comparative study among these popular gait modalities, involving the representational capacity and fusion strategy exploration, is still lacking. From the perspectives of fine vs. coarse-grained shape and whole vs. pixel-wise motion modeling, this work presents an in-depth investigation of three popular gait representations, i.e., silhouette, human parsing, and optical flow, with various fusion evaluations, and experimentally exposes their similarities and differences. Based on the obtained insights, we further develop a C2Fusion strategy, consequently building our new framework MultiGait++. C2Fusion preserves commonalities while highlighting differences to enrich the learning of gait features. To verify our findings and conclusions, extensive experiments on Gait3D, GREW, CCPG, and SUSTech1K are conducted. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Parallel Byzantine fault tolerance method for blockchain

Parallel Byzantine fault tolerance method for blockchain

引用

International Conference on Artificial Intelligence, Blockchain, Computing and Security, ICABCS 2023

作者： Dubey, Kumar Pradyot Gnanaprakasam, C.N. Khan, Ihtiram Raza Sadik, Md Shibli Al-Farhani, Liwa H. Ray, Samrat Department of Computer and Information Science Jagadguru Ram Bhadracharya Divyanga University UP Chitrakoot India Department of Electronics and Instrumentation Engineering St. Joseph’s College of Engineering Tamilnadu Chennai India Computer Science Department Jamia Hamdard Delhi India University of Portsmouth Portsmouth United Kingdom System Analysis Control and Information Processing Dep Academy of Engineering RUDN University Moscow Russia The Institute of Industrial Management Economics and Trade Peter The Great Saint Petersburg Polytechnic University Russia

ISBN: (纸本)9781032669663

Enterprise-focused blockchain applications that offer safe, anonymous, and immutable transactions are growing as blockchain technology matures. Traditional blockchain architecture has low performance and scalability, making it unsuitable for enterpriselevel applications with high concurrency and massive data. This paper studies and proposes a simplified Byzantine fault-tolerant (SBFT) consensus algorithm to improve consensus stage efficiency, a Task parallel smart contract model to maximise multi-core system parallel efficiency, and an improved blockchain system architecture to reflect the lightweight, low coupling nature of blockchain technology. It facilitates corporate application secondary development and the ParaChain blockchain and smart contract technology. Experimental verification compares the parallelization - based ParaChain blockchain’s TPS and scalability. PBFT-based blockchain systems have been greatly improved. © 2024 The Author(s).

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

A new symbol timing and carrier frequency offset estimation algorithm for OTFS for B5G 2

A new symbol timing and carrier frequency offset estimation ...

引用

2nd IEEE International Conference on Emerging Research in Computational science, ICERCS 2024

作者： Thangavel, Gunasekaran Rajan, Sheeba Subbulekshmi, S. Anthonisamy, Arun Ganapathi, Rohini Velankanni, Gowri University of Technology and Applied Sciences Department of Electronics and Communication Engineering Muscat Oman SRM Institute of Science and Technology School of Computing Tiruchirapalli India Prathyusha Engineering College Department of Electronics and Communication Engineering Chennai India Panimalar Engineering College Department of Computer Science and Business System Chennai India St. Joseph's Institute of Technology Department of Electronics and Communication Engineering Chennai India K. Ramakrishnan College of Technology Department of Electronics and Communication Engineering Trichy India

ISBN: (纸本)9798331534950

In B5G networks, accurate symbol timing estimate and estimation of the carrier frequency offset are both absolutely necessary. Particularly in highly mobile and interference-ridden settings, these factors are of paramount importance for sustaining high-throughput, dependable wireless communication. Estimating symbol timing and carrier frequency offset in OTFS can be difficult due to factors including the time-varying nature of Doppler shifts and interference from surrounding cells as well as the non-stationary nature of wireless channels. In such changing settings, traditional algorithms often fail to produce reliable estimations. To account for frequency changes caused by things like Doppler effects and drift, this research uses Machine Learning-Based Carrier Frequency Prediction Analysis (ML-CFPA) to implement a dynamic carrier frequency offset estimation and correction system. To preserve signal coherence and lower interference in B5G settings, carrier frequency offset adjustment is essential. Communication solutions with high performance are required for these applications to ensure they can function normally despite the complexity and changeability of their contexts. The outcomes demonstrate the higher performance of the algorithm in contrast to conventional approaches, highlighting its applicability to B5G networks. Its potential as a crucial enabling technology for the B5G future has been solidified by these simulations, which verify its capacity to maintain reliable communication under a variety of real-world conditions. © 2024 IEEE.

关键词： Graphical user interfaces

来源：评论

学校读者我要写书评

暂无评论

Geometry of Autometrized Lattice Ordered Monoids

arXiv

引用

arXiv 2024年

作者： Regasa, Tekalign Aklilu, Girum Venkateswarlu, Kolluru Department of Mathematics Addis Ababa University Ethiopia Department of Computer Science and System Engineering Andhra University Visakhapatnam India

In this paper, we study the geometry of Autometrized lattice ordered monoid(short as AL-monoids) by introducing the concept of metric betweeness and its properties t1,t2, B-linearity, D-linearity, lattice betweeness, B-linearity, and D-linearity, segments and equilateral triangles. It is proved that there do not exist equilateral triangles in AL-monoids. It is also proved that any AL-monoid is ptolemaic. This result subsumes all the geometry of commutative *** Codes 47H09, 47H10 © 2024, CC BY.

关键词： Algebra

来源：评论

学校读者我要写书评

暂无评论

Generation of tunable superchiral spot in metal-insulator-metal waveguide

引用

Chinese Optics Letters 2023年第1期21卷 125-130页

作者：庄涛胡海峰詹其文 School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai 200093China Zhangjiang Laboratory Shanghai 201204China Shanghai Key Laboratory of Modern Optical System University of Shanghai for Science and TechnologyShanghai 200093China

The chiral feature of an optical field can be evaluated by the parameter of g-factor enhancement,which is helpful to enhance chiroptic signals from a chiral *** this work,the superchiral spot has been theoretically proposed in metal-insulator-metal *** g-factor enhancement of the superchiral spot can be enhanced by 67-fold more than that of circularly polarized light,and the spot is confined in the deep wavelength scale along each spatial ***,the position of the superchiral spot can be tuned by manipulating the incident *** tunable superchiral spot may find applications in chiral imaging and sensing.

关键词： circular dichroism superchiral spot radially polarized beam metal-insulator-metal waveguide

来源：评论

学校读者我要写书评

暂无评论

Expediting Distributed GNN Training with Feature-only Partition and Optimized Communication Planning

Expediting Distributed GNN Training with Feature-only Partit...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Bingqian Du Jun Liu Ziyue Luo Chuan Wu Qiankun Zhang Hai Jin National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Department of Electrical and Computer Engineering The Ohio State University USA Department of Computer Science The University of Hong Kong Hong Kong School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798350383508

ISBN: (纸本)9798350383515

Feature-only partition of large graph data in distributed Graph Neural Network (GNN) training offers advantages over commonly adopted graph structure partition, such as minimal graph preprocessing cost and elimination of cross-worker subgraph sampling burdens. Nonetheless, performance bottleneck of GNN training with feature-only partitions still largely lies in the substantial communication overhead due to cross-worker feature fetching. To reduce the communication overhead and expedite distributed training, we first investigate and answer two key questions on convergence behaviors of GNN model in feature-partition based distribute GNN training: 1) As no worker holds a complete copy of each feature, can gradient exchange among workers compensate for the information loss due to incomplete local features? 2) If the answer to the first question is negative, is feature fetching in every training iteration of the GNN model necessary to ensure model convergence? Based on our theoretical findings on these questions, we derive an optimal communication plan that decides the frequency for feature fetching during the training process, taking into account bandwidth levels among workers and striking a balance between model loss and training time. Extensive evaluation demonstrates consistent results with our theoretical analysis, and the effectiveness of our proposed design.

关键词： Training Time-frequency analysis Costs Computational modeling Distributed databases Bandwidth Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Autometrized Lattice Ordered Monoids

arXiv

引用

arXiv 2024年

In this paper, we introduce the notion of Autometrized lattice ordered monoids (for short,AL-monoids) as a generalization to DRl-semi groups. We obtain the basic prop- erties of AL-monoids. Also, we prove that Autometrized lattice ordered monoids are equationally definable. Furthermore, we show that AL-monoids are an optimal common abstraction of Boolean algebras and *** Codes 06F05, 08B15 Copyright © 2024, The Authors. All rights reserved.

关键词： Boolean algebra

来源：评论

学校读者我要写书评

暂无评论

A Novel and Accurate Method of Diagnosis via AI and ML for Smart HC system 4

A Novel and Accurate Method of Diagnosis via AI and ML for S...

引用

4th IEEE International Conference on Advance Computing and Innovative Technologies in engineering, ICACITE 2024

作者： Elizabeth, J. Ruby Emmanuel, Mathews Aswini, E. Pathak, Shriram Bhuvaneswari, M. Choubey, Abhishek Nehru Institute of Engineering and Technology Department of Computer Science and Business System Tamil Nadu T.M.Palayam India Sathyabhama University of Science and Technology chennai India Prince Shri Venkateshwara Padmavathy Engineering College Department of Electronics and Communication Engineering India Ies Insititute of Technology and Management Ies University Department of Mechanical Engineering M.P. Bhopal India Karpagam Academy of Higher Education Department of Biomedical Engineering Coimbatore India Sreenidhi Institute of Science and Technology ghatkesar Hyderabad Yamnampet India

ISBN: (纸本)9798350360165

Recent breakthroughs in the internet of Issues (IoT), computing through the cloud, and the use of AI have changed the standard healthcare system, making a case for smart health care. By combining key technologies including IoT and AI, there is a major improvement in medical services. The merging of IoT and AI offers multiple possibilities within the healthcare industry. This study piece presents a new artificial intelligence- and internet of things convergence-based framework for disease detection in smart health care systems, emphasizing especially on cardiovascular diseases and diabetes. The suggested model includes separate steps, including data collection, preparation, classification, and parameter modification. Utilizing Internet of Things devices like smartwatches along with sensors enables smooth collection of information, while AI methods harness this data for the detection of illnesses. The technique presents a new approach, applying to the Crowd Search Optimizer algorithm-based Flowed Long and Short-Term Memory (CSO-CLSTM) approach to disease detection. CSO is applied to improve both 'weights' while 'bias' factors of the CLSTM framework, while the isolation Fg (iForest) method is implemented to remove outliers. Through testing with health care information, the CSO-LSTM model got excellent accuracy ratings of 96.16% when 97.26% in identifying heart disease while diabetes, respectively. Consequently, the suggested CSO-LSTM model appears as a solid and reliable method of disease evaluation in digital healthcare systems, showing its potential for broad application. © 2024 IEEE.

关键词： Electronic health record

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：