检索结果-内蒙古大学图书馆

Euromicro Conference on parallel, distributed and Network-Based Processing

作者： Massimiliano Russo Mehwish Nisar Antonio Pauciullo Pasquale Imperatore Marco Lapegna Diego Romano Dept. of Mathematics and Applications University of Naples Federico II Naples Italy Institute for the Electromagnetic Sensing of the Environment (IREA) National Research Council Naples Italy Inst. for High Performance Computing and Networking National Research Council Naples Italy

ISBN: (数字)9798331524937

ISBN: (纸本)9798331524944

Synthetic Aperture Radar (SAR) tomography is an advanced technique for monitoring deformations of the Earth’s surface. However, the computational complexity of SAR tomography algorithms often restricts their application to large-scale datasets. To address this issue, we introduce a multi-level parallel implementation of a single scatterer detection algorithm specifically designed to exploit the capabilities of modern heterogeneous High-Performance Computing (HPC) systems. By efficiently distributing the computational workload at different levels across multiple processing units, our parallel approach significantly reduces processing time, facilitating the analysis of extensive SAR datasets. We assess the performance of our parallel implementation using real-world SAR data, showcasing its effectiveness in enhancing both the efficiency and scalability of SAR tomography. Our work contributes to advancing remote sensing techniques and offers valuable insights into the application of HPC for large-scale environmental monitoring.

关键词： High performance computing Scalability Graphics processing units Tomography Performance analysis parallel algorithms Synthetic aperture radar Surface treatment Remote sensing Optimization

来源：评论

学校读者我要写书评

暂无评论

Task scheduling and communication support for parallel and distributed real-time systems

Task scheduling and communication support for parallel and d...

引用

作者： Li, Jong-Pyng Michigan State University

学位级别：Ph.D.

real-time applications are increasing in their complexity of control and computational demands. parallel and distributed systems provide cost-efficient computing power and higher degree of fault tolerance that make these systems attractive computer systems for the next generation of real-time systems. As real-time systems move from uniprocessor systems to parallel and distributed systems, the design of real-time systems becomes more complex and new techniques are required. This dissertation provides new approaches for solving two closely related problems on designing parallel and distributed real-time systems: dynamic scheduling of tasks with precedence relations and communication support for wormhole routed networks. The task scheduling algorithm combines task graph partitioning, least-laxity-first scheduling and branch-and-bound task allocation techniques to provide required real-time performance. Performance analysis show that the algorithm can efficiently schedule precedence-constrained tasks with low scheduling overhead. The parameters that affect the performance and hardware costs are studied to provide system designers with the means for fine tuning the algorithm for different system configurations. The flow control scheme of a direct network manages the network resources and directly relates to the system performance. Several flow control schemes are developed to support real-time communication on wormhole networks. The schemes differ in their priority mapping, priority adjustment, arbitration and message dropping strategies. A priority mapping scheme encodes the timing property of a message into a priority, which can be represented in a small number of digits. As the timing property of a message changes, a priority adjustment method modifies the priority to reflect the current status of the message. An arbitration function decides how to allocate bandwidth. Messages that miss their deadlines and lose their value are removed from the network by a message dro

关键词： parallel programming (Computer science)real-time data processing

来源：评论

学校读者我要写书评

暂无评论

Enhancing IoT Compliance Checking with distributed Process Mining: A Scalable Framework for Log Data Streams 18th

Enhancing IoT Compliance Checking with Distributed Process ...

引用

18th China Conference on Wireless Sensor Networks, CWSN 2024

作者： Song, Chao Ren, Zheng Hu, Ruilin Lu, Li School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9789819621859

The integration of the Internet of Things (IoT) into various industries has led to an exponential increase in the volume of data generated, posing significant challenges for compliance monitoring. Traditional compliance checking methods are often inadequate due to their inability to handle the high velocity and volume of real-time data streams emanating from IoT devices. In this paper, we introduce a novel distributed computing framework specifically designed to address these challenges by leveraging process mining techniques on IoT log data streams. The proposed framework is built to scale horizontally, allowing for the parallel processing of massive data sets while maintaining real-time performance. It is capable of identifying compliance patterns, detecting deviations, and providing actionable insights to ensure that IoT systems operate within the confines of established regulations and standards. To evaluate the performance and effectiveness of the proposed framework, we conducted extensive experiments using a publicly available industrial IoT log dataset. The results demonstrate the framework’s ability to process large volumes of data in real-time, offering a significant improvement over existing centralized approaches. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Data streams

来源：评论

学校读者我要写书评

暂无评论

A Framework for the Development of parallel and distributed real-time Embedded systems

A Framework for the Development of Parallel and Distributed ...

引用

38th EUROMICRO Conference on Software Engineering and Advanced Applications (SEAA) / 15th EUROMICRO Digital systems Design (DSD)

作者： Garibay-Martinez, Ricardo Ferreira, Luis Lino Pinho, Luis Miguel Polytech Inst Porto CISTER INESC TEC ISEP Oporto Portugal

ISBN: (纸本)9780769547909

Embedded real-time applications increasingly present high computation requirements, which need to be completed within specific deadlines. But, those applications present highly variable patterns, depending on the data set in a determined instant. The current trend to provide parallel processing in the embedded domain allows providing higher processing power;however, it does not address the variability in the processing pattern. Dimensioning each device for its worst-case scenario implies lower average utilization, and increased available, but unusable, processing in the overall system. A solution for this problem is to extend the parallel execution of the applications, allowing networked nodes to distribute the workload, on peak situations, to neighbour nodes. In this context, this paper proposes a framework to develop parallel and distributed real-time embedded applications, transparently using OpenMP and Message Passing Interface (MPI), within a programming model based on OpenMP. The paper also devises an integrated timing model, which enables the structured reasoning on the timing behaviour of these hybrid architectures.

关键词： real-time parallel execution distributed embedded systems hybrid programming model OpenMP MPI

来源：评论

学校读者我要写书评

暂无评论

parallel and distributed real-time systems

引用

JOURNAL OF systems AND SOFTWARE 2005年第1期77卷 1-2页

作者： Manimaran, G Ecker, K Huh, EN Iowa State Univ Dept Elect & Comp Engn Ames IA 50011 USA Tech Univ Clausthal D-3392 Clausthal Zellerfeld Germany Sahmyook Univ Seoul South Korea

No abstract available

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic Bio-Information Image Recognition of Cancel Cell Death based on parallel Multiple Artificial Intelligence Modules

Dynamic Bio-Information Image Recognition of Cancel Cell Dea...

引用

International Conference on Advanced Communication Technology (ICACT)

作者： Ming-Shen Jian Yi-Ling Ye Wen-Yu Chung Wen-Xuan Li Yan-Ting Lin Bo-Yuan Huang Javeria Sikandar Yu-Jung Chang Dept. of CSIE CCIS Lab National Formosa University Yunlin Taiwan Dept. of Biotechnology National Formosa University Yunlin Taiwan

ISBN: (数字)9791188428137

ISBN: (纸本)9798331507602

This investigation purpose a unique Dynamic Bio-Information Image Recognition System (DBIRS) built for the real-time monitoring of cancer cell apoptosis produced by available medical therapy. Utilizing advances in microscopy, AI-based image recognition, and parallel computing across distributed edge devices, the proposed system dramatically boosts the precision and efficiency of cellular investigation. By deploying numerous AI modules adapted to distinct cellular states, the system provides automatic, high-accuracy recognition of morphological changes in cancer cells, hence minimizing the need on manual procedures. This technique offers remote monitoring and fast data processing via cloud integration, permitting continuous observation and timely intervention. The system's efficiency is proven through trials on K-562 cells, where it attained a recognition accuracy of 97.41%. These outcomes underline the potential of the proposed method to refine cytotoxicity assays, delivering a strong tool for increasing the efficacy and safety of immune cell treatments in clinical and research settings.

关键词： Training Image recognition Accuracy real-time systems Safety Artificial intelligence Biological cells Remote monitoring Biomedical imaging Cancer

来源：评论

学校读者我要写书评

暂无评论

Energy-minimized Scheduling of real-time parallel Workflows on Heterogeneous distributed Computing systems

引用

IEEE TRANSACTIONS ON SERVICES COMPUTING 2022年第5期15卷 2766-2779页

作者： Hu, Biao Cao, Zhengcai Zhou, MengChu Beijing Univ Chem Technol Coll Informat Sci & Technol Beijing 100029 Peoples R China New Jersey Inst Technol Dept Elect & Comp Engn Newark NJ 07102 USA King Abdulaziz Univ Ctr Res Excellence Renewable Energy & Power Syst Jeddah 21589 Saudi Arabia

Today's large-scale parallel workflows are often processed on heterogeneous distributed computing platforms. From an economic perspective, computing resource providers should minimize the cost while offering high service quality. It has become well-recognized that energy consumption accounts for a large part of a computing system's total cost, and timeliness and reliability are two important service indicators. This work studies the problem of scheduling a parallel workflow that minimizes the system energy consumption under the constraints of response time and reliability. We first mathematically formulate this problem as a Non-linear Mixed Integer Programming problem. Since this problem is hard to solve directly, we present some highly-efficient heuristic solutions. Specifically, we first develop an algorithm that minimizes the schedule length while meeting reliability requirement, on top of which we propose a processor-merging algorithm and a slack time reclamation algorithm using a dynamic voltage frequency scaling (DVFS) technique to reduce energy consumption. The processor-merging algorithm tries to turn off some energy-inefficient processors such that energy consumption can be minimized. The DVFS technique is applied to scale down the processor frequency at both processor and task levels to reduce energy consumption. Experimental results on two real-life workflows and extensive synthetic parallel workflows demonstrate their effectiveness.

关键词： Reliability Task analysis Energy consumption Program processors time factors Schedules Heuristic algorithms Dynamic voltage frequency scaling heuristics non-linear mixed integer programming optimization reliability response time workflow scheduling

来源：评论

学校读者我要写书评

暂无评论

Open real-time environment for parallel and distributed systems

Open real-time environment for parallel and distributed syst...

引用

20th International Conference on distributed Computing systems (ICDCS 2000)

作者： Kuo, Tei-Wei Lin, Kwei-Jay Wang, Yu-Chung Natl Chung Cheng Univ Chiayi Taiwan

Most computer-based systems have hard real-time constraints. Schedulers in complex systems must be designed to manage a set of applications developed and deployed independently. In this paper, we study an open real-time environment architecture for distributed systems where real-time applications may run concurrently with non-real-time applications. The architecture uses a two-level scheduling scheme. Each application is assigned a sporadic server to schedule the processes in the application. All sporadic servers are then scheduled by a system-wide fixed priority scheduler. Using the proposed open environment architecture, all hard real-time applications are guaranteed to have their reserved CPU utilization in order to meet all their deadlines. The guarantee is independent of the behaviors of all other applications in the same system. We present the schedulability analysis methods on systems with or without shared memory.

关键词： real time systems

来源：评论

学校读者我要写书评

暂无评论

Multi-Node Inference Architectures for Low-Latency LLM Serving

Multi-Node Inference Architectures for Low-Latency LLM Servi...

引用

Advanced Computing Technologies (ICoACT), International Conference on

作者： Naresh Kumar Gundla Sri Harsha Atthuluri

ISBN: (数字)9798331509859

ISBN: (纸本)9798331509866

Over the past few years, large language models have evolved to enable a wide range of applications-from natural language understanding to real-time conversational agents. However, the deployment of LLMs into production presents many significant challenges, especially with regard to low-latency responses that enable real-time interactions. This work investigates multi-node inference architectures for optimized deployment using open-source frameworks with scalability, flexibility, and cost-effectiveness. We investigate various methods, such as microbatching, tensor and pipeline parallelism, and sophisticated load balancing, that effectively distribute inference workloads across multiple nodes. We conduct extensive evaluations using popular open-source tools such as Kubernetes, Ray, and Envoy to benchmark the performance of these architectures in terms of latency, throughput, and resource utilization under diverse workloads. We also analyze model replication versus model partitioning trade-offs, giving insights into the most appropriate configuration for various deployment scenarios. As our results show, a well-orchestrated multi-node setup can be used to greatly reduce inference latency while preserving high throughputs, enabling the deployment of sophisticated LLMs in latencysensitive applications. This paper gives insights with a detailed analysis of multi-node inference strategies and integration into open-source ecosystems, therefore it will be a great guide for practitioners seeking to develop deployments of LLMs at scale. In summary, this work underlines how distributed architectures can overcome some of the inherent limitations imposed by singlenode deployments and are crucial for achieving more efficient and responsive AI-driven services.

关键词： Tensors Large language models Scalability Pipelines Computer architecture parallel processing Throughput real-time systems Low latency communication Load modeling

来源：评论

学校读者我要写书评

暂无评论

Publicly Verifiable distributed Computation for MEC Setting

引用

IEEE Transactions on parallel and distributed systems 2025年第7期36卷 1416-1430页

作者： Wang, Qiang Li, Zhicheng Zhou, Fucai Xu, Jian Zhang, Changsheng Northeastern University Software College Shenyang110013 China

With the rapid expansion of the Internet of Things (IoT), the shift from cloud computing to Mobile Edge Computing (MEC) has become necessary to address the low-latency requirements of real-time applications. Verifiable computation (VC) enables resource-limited clients to outsource their computation-intensive tasks to a powerful cloud while ensuring the correctness of the computation result. However, traditional VC schemes, originally designed for cloud computing, face challenges when applied to MEC environments, such as scalability issues, robustness, and efficiency concerns. To this end, we propose a verifiable distributed computation scheme for MEC, where computation tasks are distributed between a cloud server cluster (consisting of n servers) and an edge server. The cloud handles most of the computation through parallel sub-tasks, while the edge server verifies intermediate results and performs minimal computation to recover the final outcome. Our scheme guarantees that the result can be recovered if at least t servers, out of a total of n servers in the cloud server cluster, perform their computations honestly. By leveraging batch verification and matrix-optimized polynomial evaluations, our scheme significantly enhances scalability, fault tolerance, and efficiency. The extensive analysis and simulations demonstrate that our proposed scheme is more feasible than existing solutions. © 1990-2012 IEEE.

关键词： distributed cloud

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：