检索结果-内蒙古大学图书馆

IEEE Conference on Vehicular Technology (VTC)

作者： Abhijeet Manoj Varma Nikumani Choudhury Jay Dave Anakhi Hazarika Moustafa M. Nasralla Dept. of Computer Science & Information Systems Birla Institute of Technology & Science Pilani Hyderabad India Dept. of Electrical & Electronics Engineering Birla Institute of Technology & Science Pilani Hyderabad India Dept. of Communications & Networks Engineering Smart Systems Engineering Lab Prince Sultan University Riyadh Saudi Arabia

ISBN: (数字)9798350387414

ISBN: (纸本)9798350387421

As the Internet of Things (IoT) is poised to become a global phenomenon, it is imperative to schedule the transmissions of IoT devices effectively and in a fair way. Leveraging Long Range (LoRa) technology, we can achieve transmissions that consume minimal power while covering vast distances, aligning with the requirements of IoT devices. However, the proximity of multiple devices within the same area often leads to packet interference and collisions. To address this, our study introduces a pioneering scheduling method utilizing a constellation of Low Earth Orbit (LEO) satellites to manage and streamline the transmission of data from End Devices (EDs). This method employs two LEO satellites: the first satellite assigns the sequence for EDs to dispatch their packets, and the second collects these packets in the predetermined sequence before forwarding them to the LoRa Network Server (LNS). For urgent (URG) communications, EDs can alert the first satellite, which then coordinates with the LNS to schedule these priority transmissions. The LNS generates a schedule that is relayed to the second satellite, informing EDs with URG packets of their specific transmission times and channels. This scheduling approach is designed to optimize channel usage effectively while accommodating the transmission of urgent data.

关键词： 6G mobile communication Schedules Vehicular and wireless technologies Satellites scheduling algorithms Low earth orbit satellites LoRaWAN

来源：评论

学校读者我要写书评

暂无评论

Towards Fair and Firm Real-Time scheduling in DNN Multi-Tenant Multi-Accelerator Systems via Reinforcement Learning

Towards Fair and Firm Real-Time Scheduling in DNN Multi-Tena...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Enrico Russo Francesco Giulio Blanco Maurizio Palesi Giuseppe Ascia Davide Patti Vincenzo Catania Department of Electrical Electronic and Computer Engineering (DIEEI) University of Catania Catania Italy

ISBN: (数字)9798350330991

ISBN: (纸本)9798350331004

This paper addresses the critical challenge of managing Quality of Service (QoS) in cloud services, focusing on the nuances of individual tenant expectations and varying Service Level Indicators (SLIs). It introduces a novel approach utilizing Deep Reinforcement Learning for tenant-specific QoS management in multi-tenant, multi-accelerator cloud environments. The chosen SLI, deadline hit rate, allows clients to tailor QoS for each service request. A novel online scheduling algorithm for Deep Neural Networks in multi-accelerator systems is proposed, with a focus on guaranteeing tenant-wise, model-specific QoS levels while considering real-time constraints.

关键词： Feedback loop scheduling algorithms Circuits and systems Focusing Quality of service Artificial neural networks Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Criticality-Aware Load-Balanced scheduling Algorithm for Teleoperated Vehicles Controlled over Vehicle-to-Infrastructure Network

Criticality-Aware Load-Balanced Scheduling Algorithm for Tel...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Subhendu Bhadra Sangyoung Park Smart Mobility Systems Technical University of Berlin

ISBN: (数字)9798331517786

ISBN: (纸本)9798331517793

Teleoperation is drawing attention as a complementary technology for autonomous driving to ensure a safe and reliable transportation. Human teleoperators making decisions based on real-time videos may serve as a fallback solution for corner cases where autonomous driving algorithms do not operate well. However, teleoperation comes with its own challenges such as limited network bandwidth and latency of Vehicular Ad Hoc Network (VANET). In this paper, we investigate a scenario comprising multiple teleoperated vehicles under such network circumstances. The real-time videos from the vehicle to the teleoperator and the commands are both subject to real-time constraints. We propose a Criticality-Aware Load-Balanced scheduling for Vehicles (V-CALBS) algorithm with VANET management and packet scheduling, capable of greatly increasing Time-To-Collision (TTC) and thus enhancing the safety of the vehicles and the reliability of the traffic.

关键词： Teleoperators Vehicular and wireless technologies scheduling algorithms Vehicular ad hoc networks Real-time systems scheduling Safety Reliability Autonomous vehicles Videos

来源：评论

学校读者我要写书评

暂无评论

AI-Powered Federated Task scheduling and Self-Healing Framework in Dynamic Cloud Systems

AI-Powered Federated Task Scheduling and Self-Healing Framew...

引用

IEEE International Conference on Utility and Cloud Computing (UCC)

作者： Umit Demirbaga Omer Rana Ashiq Anjum Gagangeet Singh Aujla Department of Computer Engineering Bartin University Türkiye Department of Computer Science and Informatics Cardiff University United Kingdom Department of Computing and Mathematical Sciences University of Leicester United Kingdom Department of Computer Science Durham University United Kingdom

ISBN: (数字)9798350367201

ISBN: (纸本)9798350367218

Federated cloud environments have emerged to integrate multiple cloud providers like AWS, Azure, and Google Cloud seamlessly into cloud computing. Optimising resource utilisation and ensuring high availability in such environments pose significant challenges. This paper comprehensively investigates federated task scheduling algorithms and self-healing mechanisms in autonomous federated cloud setups. The research objectives include the development of an independent task-scheduling algorithm capable of intelligently distributing computing tasks across federated clouds based on workload characteristics, resource availability, and network latency. Furthermore, the study investigates implementing self-healing mechanisms to detect faults and performance degradation, triggering automatic recovery processes for uninterrupted service availability. The proposed approaches are evaluated through real-world experiments, considering diverse cloud workloads and failure scenarios, focusing on resource utilisation efficiency, system performance, and the effectiveness of the self-healing mechanisms in mitigating cloud failures and maintaining seamless operations within the federated environment.

关键词： Degradation Cloud computing scheduling algorithms Federated learning System performance Scalability Focusing Dynamic scheduling Real-time systems Resource management

来源：评论

学校读者我要写书评

暂无评论

A Review on Two-stage Robust Optimal scheduling of Virtual Power Plants Containing Wind Power Integration

A Review on Two-stage Robust Optimal Scheduling of Virtual P...

引用

IEEE Conference on Energy Internet and Energy System Integration (EI2)

作者： Zitong Huang Yangjue Huang Zhipeng Huang Electric Power Research Institute of Guangdong Power Grid Co. Ltd. Guangzhou China Guangdong Provincial Key Laboratory of Electric Power Equipment Reliability

ISBN: (数字)9798331523527

ISBN: (纸本)9798331523534

With the development of my country's power industry and the needs of economic growth, the consumption of fossil energy by the power system is increasing. Natural fossil energy reserves such as coal are declining year by year and the burning of a lot of fossil energy will cause air pollution. As a clean and renewable energy source, wind power is occupying an increasingly important position in China, however, the integration of wind power will have a bad affect to the power system. The operation mode of the virtual power plant can provide good operating conditions for the grid connection of wind turbines, realize the efficient use of wind energy, and reduce the wind curtailment rate caused by grid connection. Due to the strong randomness and intermittence of wind power [1], it is a problem that needs to be studied to optimize the scheduling of virtual power plants with grid-connected wind power. This paper first constructs the structure of the virtual power plant, including thermal power and hydropower units, interruptible loads, energy storage equipment and wind power units, establishes objective functions and constraints, and builds a two-stage robust optimal dispatch model. In response to the uncertainty of wind power output and load supply, the formulated dispatching plan has better operating economy. In order to solve the min-max-min two-stage robust model, a column constraint generation algorithm is used to divide the model into a main problem and a sub-problem, then use strong duality theory to convert the subproblem into a single-layer solution to the maximum value problem, and use the big-M algorithm to linearize the subproblem. Finally, an example is given to verify the economics of the two-stage robust optimization scheduling model scheduling scheme, which can greatly reduce the scheduling operating cost compared with the deterministic optimal scheduling scheme. Then the effects of different robust adjustment parameters and wind power output fluctuation val

关键词： Uncertainty Wind energy scheduling algorithms Optimal scheduling System integration Wind power generation Virtual power plants Wind turbines Thermal loading Load modeling

来源：评论

学校读者我要写书评

暂无评论

Design of Big Data scheduling Optimization Algorithm Based on Deep Reinforcement Learning

Design of Big Data Scheduling Optimization Algorithm Based o...

引用

Telecommunications and Power Electronics (TELEPE), International Conference on

作者： Xiao Chen Guangdong Polytechnic of Science and Technology Zhuhai Guangdong China

ISBN: (数字)9798350369212

ISBN: (纸本)9798350380002

With the advent of the network era, data processing and optimization has become a key issue. Under this background, the research aims to study the big data scheduling optimization algorithm based on Deep Reinforcement Learning (DRL) to improve the efficiency and quality of data processing. In this article, an optimization algorithm of big data scheduling based on DRL is proposed. Through DRL technology, computers can schedule computing tasks and allocate resources independently according to historical data and real-time information, thus realizing more intelligent data processing. The research results show that the DRL-based big data scheduling optimization algorithm can significantly improve computing efficiency, reduce resource waste and optimize task scheduling when dealing with large-scale data. The big data scheduling optimization algorithm based on DRL has obvious advantages in running time, convergence and scheduling accuracy, and provides a more efficient and intelligent data processing method for practical application fields.

关键词： Accuracy Processor scheduling scheduling algorithms Big Data Data processing Deep reinforcement learning scheduling Telecommunications Optimization Convergence

来源：评论

学校读者我要写书评

暂无评论

Distribution Network Optimal scheduling Model Considering Charge-Discharge Characteristics of Electricity-Hydrogen Storage

Distribution Network Optimal Scheduling Model Considering Ch...

引用

Energy Engineering and Power Systems (EEPS), International Conference on

作者： Jianhua Liu Yong Zhang Han Zhou Xin Yu Ruxin Bai State Grid Liaoning Electric Power Co. Ltd. Dandong Power Supply Company Dandong China

ISBN: (数字)9798350375138

ISBN: (纸本)9798350375145

The power supply and demand imbalance caused by randomness is always an important problem to be solved in the operation. Effectively improve the operation economy of distribution network and realize the energy scheduling capability of distribution network. Firstly, t he charge-discharge characteristics of electricity-hydrogen storage system are studied in this paper. By analyzing related technical parameters and characteristics, the model of electric hydrogen system is established. Secondly, taking the minimization of the load peak and valley difference and the minimization of the objectives, the rational scheduling are realized by optimizing the scheduling algorithm on the basis of considering the randomness factors. A distribution network optimal scheduling model considering the characteristics of electricity-hydrogen storage is established. Finally, the optimum operation of the distribution system before and after taking into account properties is simulated. It is proved that the optimal scheduling model is accurate and superior with respect to the charging and discharging characteristics of hydrogen storage. In summary, the optimal distribution network scheduling model proposed. It has certain scalability. In the future, we can further study and explore the influence further improve and optimize the model to achieve higher economy and reliability.

关键词： Analytical models scheduling algorithms Hydrogen Optimal scheduling Distribution networks Power system stability Minimization Stability analysis Sustainable development Load modeling

来源：评论

学校读者我要写书评

暂无评论

Leveraging Multi-Instance Gpus Through Moldable Task scheduling

SSRN

引用

SSRN 2024年

作者： Villarrubia, Jorge Costero, Luis Igual, Francisco Daniel Olcoz, Katzalin Universidad Complutense de Madrid Departamento de Arquitectura de Computadores y Automática Madrid28040 Spain

NVIDIA MIG (Multi-Instance GPU) allows partitioning a physical GPU into multiple logical instances with fewer fully-isolated resources, which can be dynamically reconfigured. We discuss the opportunities and challenges that a joint dynamic MIG reconfiguration and a task scheduling algorithm pose on modern GPUs. Specifically, we propose a makespan minimization problem for multi-task execution under the constraints imposed by the MIG technology on Ampere (NVIDIA A30/A100) and Hopper (H100) GPUs. We justify that, contrary to traditional scheduling algorithms targeting multi-cores, task monotony should not be assumed, and we present an algorithm (FAR) to solve the problem without that assumption. FAR schedules tasks in batches offline, and their schedules are concatenated on the fly in an improved way to favor resource reuse. We mathematically prove that the algorithm is a 7/4-approximation for the A30 and a 2-approximation for the A100/H100, with twice the competitive ratio for multi-batch scheduling. However, our experiments in real-world scenarios reveal that the makespan does not exceed the optimum by more than 1.22× for a well-known suite of benchmarks, and 1.10× for synthetic inputs inspired by real kernels. We obtain good experimental results for each batch of tasks, but also in the concatenation of batches, with large improvements over the state of the art and proposals that do not exploit the GPU reconfiguration. Moreover, we show that the heuristics used allow a correct adaptation to tasks of very different characteristics. © 2024, The Authors. All rights reserved.

关键词： scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Research on Multi-direction N-RGVs (N-Rail Guided Vehicles) scheduling System Based on Bubble Sort Algorithm

Research on Multi-direction N-RGVs (N-Rail Guided Vehicles) ...

引用

Robotics, Intelligent Control and Artificial Intelligence (RICAI), International Conference on

作者： Yuxuan Han Jianqiao Xiong Zongrong Wang Ming Li Yifan Wu School of Mechanical Engineering Nanjing Institute of Technology Nanjing Institute of Technology NJIT Nanjing China School of Mechanical Engineering Nantong Institute of Technology Nantong Institute of Technology NIT Nantong China

ISBN: (数字)9798331541699

ISBN: (纸本)9798331541705

To address issues related to queuing of four-way vehicles and potential collisions, a novel bubble sort scheduling algorithm was designed with the aim of enhancing the operational efficiency of four-way vehicles. Based on the bubble sort algorithm, with the objective of minimizing the total time for goods to be shipped out of the warehouse, and with the constraint of multiple four-way vehicles (N-RGVs) operating together, a multi-closed loop system model for inbound and outbound warehouse operations was constructed. In this scheduling system, a novel bubble sort scheduling algorithm was introduced to perform a new sorting and path planning for multiple four-way vehicles. In comparison with the A* scheduling algorithm and the closed-loop scheduling algorithm, extensive tests were conducted on an experimental platform to gather relevant data for the three algorithms. The experimental results indicate that the novel bubble sort scheduling algorithm effectively resolves issues of queuing and vehicle conflicts, reduces computational costs, and thus significantly enhances overall system efficiency.

关键词： scheduling algorithms scheduling Path planning Computational efficiency Artificial intelligence Robots Sorting Intelligent control

来源：评论

学校读者我要写书评

暂无评论

FLUS-FoV: User scheduling for Federated Learning-Based 360-Degree Video FoV Prediction

FLUS-FoV: User Scheduling for Federated Learning-Based 360-D...

引用

International Conference on Computers, Communications, and Systems (ICCCS)

作者： Feng Wen Jie Li Hefei University of Technology HeFei China

ISBN: (数字)9798350350210

ISBN: (纸本)9798350350227

The accurate prediction of the Field of View (FoV) is instrumental in optimizing the streaming performance of 360-degree videos, considering that users typically focus on specific portions of the video (denoted as FoV)and periodically switch between them. However, when privacy concerns restrict the sharing of user view data, the inability to utilize this data for training network models significantly hampers the precision of FoV prediction. To tackle this challenge, this paper introduces an innovative federated learning-based FoV prediction algorithm. This algorithm harnesses a combination of 360-degree video saliency and user view trajectories to forecast FoV with precision. Furthermore, it presents a novel algorithm for quantifying the information value with communication resource allocation to optimize this prediction process. Through experimental assessments, our findings demonstrate that the proposed scheduling algorithm outperforms traditional methods, and the distributed FoV prediction accuracy closely approximates centralized approaches.

关键词： Training Privacy Accuracy scheduling algorithms Switches Prediction algorithms Trajectory

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：