检索结果-内蒙古大学图书馆

IEEE Conference on Vehicular Technology (VTC)

作者： Tanniru Abhinav Siddharth Kethu Sesha Sarath Reddy Joseph John Cherukara Deepak Gangadharan Computer Systems Group IIIT Hyderabad

ISBN: (数字)9798331517786

ISBN: (纸本)9798331517793

With advances in vehicular communications technology such as Vehicle-to-Infrastructure (V2I) and Vehicle-to-Vehicle (V2V), the computation of vehicular tasks has become distributed facilitated by computation offloading from the vehicular platform to road side units (RSUs) or the cloud infrastructure. In this work, due to communication overheads, we do not consider the cloud and only consider offloading horizontally across RSU nodes. However, the advantage of offloading depends upon how the tasks are scheduled across the RSUs. Unlike existing horizontal offloading works, this work explores the benefits of online task partitioned scheduling for computation offloading from multiple vehicles to RSUs or edge nodes. Specifically, we propose a new efficient timeslot-based online hybrid partitioned scheduling algorithm, which splits some tasks into subtasks and schedules them across RSU nodes while considering vehicle flow constraints. We compared and evaluated the effectiveness of our proposed hybrid partitioned scheduling algorithm with the fully partitioned algorithm. We also compared the performance of the aforementioned algorithms with an optimal scheduling algorithm utilizing several experiments conducted on a real-world vehicular dataset.

关键词： Vehicular and wireless technologies Schedules scheduling algorithms Heuristic algorithms Vehicle-to-infrastructure Road side unit Optimal scheduling Dynamic scheduling Partitioning algorithms Edge computing

来源：评论

学校读者我要写书评

暂无评论

Fully Distributed EV Charging scheduling for Load Flattening in V2G Systems

Fully Distributed EV Charging Scheduling for Load Flattening...

引用

IEEE Conference on Decision and Control

作者： Jinwook Heo Soojeong Hyeon Hyungbo Shim Jinsung Kim Department of Electrical and Computer Engineering ASRI Seoul National University Seoul Korea Vehicle SW Division Hyundai Motor Company Hwaseong Korea

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

This paper presents a fully distributed algorithm for scheduling electric vehicle (EV) charging and discharging to flatten the total load of the grid, while considering constraints on grid transmission capacity. As a fully distributed solution, the proposed algorithm operates without the need for a central unit. Instead, each agent only communicates a single dual variable with its neighboring agents based on a communication graph, and thus no private information is shared. In particular, the algorithm does not rely on initial conditions, ensuring robustness in online changes of operational conditions. Simulation results verify the effectiveness of the proposed algorithm.

关键词： Vehicle-to-grid Privacy scheduling algorithms Heuristic algorithms Simulation scheduling Electric vehicle charging Robustness Vehicle dynamics Distributed algorithms

来源：评论

学校读者我要写书评

暂无评论

Flag-Proxy Networks: Overcoming the Architectural, scheduling and Decoding Obstacles of Quantum LDPC Codes

Flag-Proxy Networks: Overcoming the Architectural, Schedulin...

引用

IEEE/ACM International Symposium on Microarchitecture (MICRO)

作者： Suhas Vittal Ali Javadi-Abhari Andrew W. Cross Lev S. Bishop Moinuddin Qureshi Georgia Institute of Technology Atlanta GA United States IBM T.J. Watson Research Center Yorktown Heights NY United States

ISBN: (数字)9798350350579

ISBN: (纸本)9798350350586

Quantum error correction is necessary for achieving exponential speedups on important applications. The planar surface code has remained the most studied error-correcting code for the last two decades because of its relative simplicity. However, encoding a singular logical qubit with the planar surface code requires physical qubits quadratic in the code distance $(d)$ , making it space-inefficient for the large-distance codes necessary for promising applications. Thus, Quantum Low-Density Parity-Check (QLDPC) have emerged as an alternative to the planar surface code but require a higher degree of connectivity. Furthermore, the problems of fault-tolerant syndrome extraction and decoding are understudied for these codes and also remain obstacles to their usage. In this paper, we consider two under-studied families of QLDPC codes: hyperbolic surface codes and hyperbolic color codes. We tackle the three challenges mentioned above as follows. First, we propose Flag-Proxy Networks (FPNs), a generalizable architecture for quantum codes that achieves low connectivity through flag and proxy qubits. Second, we propose a greedy syndrome extraction scheduling algorithm for general quantum codes and further use this algorithm for fault-tolerant syndrome extraction on FPNs. Third, we present two decoders that leverage flag measurements to decode the hyperbolic codes accurately. Our work finds that degree-4 FPNs of the hyperbolic surface and color codes are respectively $2.9\times$ and $5.5\times$ more space-efficient than the $d=5$ planar surface code, and become even more space-efficient when considering higher distances. The hyperbolic codes also have error rates comparable to their planar counterparts.

关键词： Fault tolerance Codes Protocols scheduling algorithms Error analysis Qubit Fault tolerant systems Color Parity check codes Decoding

来源：评论

学校读者我要写书评

暂无评论

Stream-K++: Adaptive GPU GEMM Kernel scheduling and Selection using Bloom Filters

arXiv

引用

arXiv 2024年

作者： Sadasivan, Harisankar Osama, Muhammad Podkorytov, Maksim Huang, Carlus Liu, Jun Advanced Micro Devices Inc. Bellevue United States Advanced Micro Devices Inc. Santa Clara United States Advanced Micro Devices Inc. Shanghai China

General matrix multiplication (GEMM) operations are crucial in various computational fields. As GPU architectures evolve, optimizing GEMM performance becomes increasingly important. This paper introduces Stream-K++, an enhancement to the promising Stream-K GEMM scheduling algorithm. We expand Stream-K’s scheduling policies from three to seven and implement an efficient solution selection mechanism using Bloom filters. Our approach rapidly eliminates up to ∼95.8% of unsuitable configurations while maintaining a 100% true-negative rate. Implemented using the AMD Composable Kernel library and evaluated on AMD Instinct™ MI250X GPUs, Stream-K++ demonstrates significant performance gains (up to ∼43%) in select scenarios. It remains competitive (within 20% of optimal) for ∼60-97.6% of problem sizes. Our flexible framework, implemented in the Opensieve C++ library, allows for easy adaptation to new problem sizes, scheduling policies, or additional tuning parameters, paving the way for future optimizations in GPU-based GEMM operations. © 2024, CC BY-SA.

关键词： scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Research on Optimization scheduling Algorithm of Engineering Project Resources Based on Cloud Computing

Research on Optimization Scheduling Algorithm of Engineering...

引用

Sensors, Electronics and Computer Engineering (ICSECE), IEEE International Conference on

作者： Ping Dong Cangzhou Jiaotong College Huanghua China

ISBN: (数字)9798350373646

ISBN: (纸本)9798350373653

Aiming at the problem of resource optimization scheduling for large and complex engineering projects, this paper proposes an optimization scheduling algorithm system based on cloud computing, which deeply integrates cloud computing, convolutional neural network (CNN) and Beetle Antennae Search (BAS). The Cloud Convolution Antennae Search (CCAS) model is proposed to realize efficient, accurate and dynamic resource allocation and scheduling. With cloud computing as the underlying support, the CCAS model realizes efficient storage, processing and real-time update of large-scale engineering project data through the powerful computing power and elastic expansion characteristics of cloud computing, ensuring that the model can still maintain efficient operation in the face of massive and dynamic data, and providing a solid foundation for subsequent intelligent decision-making. Secondly, at the algorithmic level, CNN is applied to the prediction of resource demand of engineering projects. It uses its advantages in spatiotemporal series data modeling to carry out deep learning on multi-dimensional data such as project schedule and resource consumption to achieve accurate prediction of future resource demand. At the same time, a Longniu whisker search algorithm is introduced to make resource scheduling decision. BAS simulates the natural process of Longurus searching for food, explores the scheduling space effectively by iteratively updating the search path, and finds the global optimal or near-optimal scheduling scheme. The organic combination of the three not only provides strong computing power support and the advantages of data management and deep learning in complex data mining, but also utilizes the efficiency of optimization algorithms in large-scale problem solving. In terms of system architecture, the three-layer cloud service model based on IaaS, PaaS and SaaS is built to provide comprehensive support for the research and development, deployment, operation and appli

关键词： Cloud computing Job shop scheduling scheduling algorithms Computational modeling Prediction algorithms scheduling Data models Convolutional neural networks Optimization Antennas

来源：评论

学校读者我要写书评

暂无评论

Resource Optimization in Cloud Environment Using Advanced Metaheuristic scheduling Algorithm

Resource Optimization in Cloud Environment Using Advanced Me...

引用

Sustainable Communication Networks and Application (ICSCNA), International Conference on

作者： Pasnur Deeplaxmi Cholleti Vikas Santhosh Kumar Medishetti Geetanjali College of Engineering and Technology Hyderabad T.S. India Keshav Memorial Institute of Technology Hyderabad T.S. India Nalla Narasimha Reddy Education Society's Group of Institutions Hyderabad T.S. India

ISBN: (数字)9798331530013

ISBN: (纸本)9798331530020

Task scheduling (TS) especially in cloud environments is quite a challenge because the resources within cloud are dynamic while the tasks present within a cloud have diverse characteristics. This paper aims at presenting a new scheduling technique based on one of the metaheuristic algorithms called Harris Hawks Optimization (HHO), which is derived from the social instinct and predatory hunting behavior of the Harris's hawks. The HHO algorithm is intended to deal with the challenges that appear in the pattern of tasks by adjusting the important factors like makespan, energy and resource utilization. We evaluate the performance of HHO through comprehensive simulations and compare it against traditional scheduling methods, highlighting its efficacy in balancing load and minimizing latency. The results demonstrate that the HHO-based scheduling approach achieves notable improvements in scheduling efficiency, with 15% improvement in energy consumption, a 16% reduction in makespan, and an 18% increase in throughput. The work that is presented in the paper offers a comprehensive approach to effective coordination of tasks in cloud computing systems to propose a satisfactory way to address this challenge in the context of current application-driven compute-intensive applications.

关键词： Cloud computing Energy consumption scheduling algorithms Heuristic algorithms Green communications Metaheuristics Throughput Dynamic scheduling Resource management Load modeling

来源：评论

学校读者我要写书评

暂无评论

An Empirical Design and Implementation of Job scheduling Enhancement for Kubernetes Clusters

An Empirical Design and Implementation of Job Scheduling Enh...

引用

International Conference on Information Networking

作者： Van-Binh Duong Duong Phung-Ha Jangwon Lee Younghan Kim School of Electronic Engineering Soongsil University Seoul Korea

ISBN: (数字)9798350330946

ISBN: (纸本)9798350330953

Batch job scheduling is a popular method in scheduling topics achieving considerable results. However, it is novel to bring those achievements to the cloud where problems of environments would limit algorithms. Although some projects have been introduced recently with benefits directly targeting machine learning jobs in cloud-native environments, there are gaps to be fulfilled. Consequently, this paper proposes an empirical design and implementation of deadline-aware enhancement of job scheduling for cloud-native environments. The proposal’s target is to automatically monitor, provision, and maintain machine learning jobs for batch job scheduling in cloud-native environments. The performance evaluation shows that the proposal has succeeded in reducing the average response time of machine learning jobs scheduled by different algorithms.

关键词： Performance evaluation Machine learning algorithms scheduling algorithms Clustering algorithms Machine learning scheduling Proposals

来源：评论

学校读者我要写书评

暂无评论

Online Multi-appointment scheduling under Demand for Oncology Patient Treatment

Online Multi-appointment scheduling under Demand for Oncolog...

引用

International Conference on Control, Decision and Information Technologies (CoDIT)

作者： Macarena Fredes Sebastián Dávila Franco Quezada MSc Industrial Engineering (MCII) Program at the Industrial Engineering Department Universidad de Santiago de Chile (USACH) Industrial Engineering Department Universidad de Santiago de Chile (USACH) Laboratory for the Development of Sustainable Production Systems (LDSPS) Chile

ISBN: (数字)9798350373974

ISBN: (纸本)9798350373981

This research is dedicated to formulating a practical problem related to the assignment of oncology clinic routes. The objective is to schedule medical events for patients attending health centers, taking into account the variable demand and limited resources in this field. To achieve this, three types of solutions are considered. The first is an Online scheduling Algorithm, which simulates a scheduling process based on medical routes and also includes its inverse version. The next is a Mixed-Integer Linear Programming (MILP) model. In this case, information is centralized, and patients are scheduled based on their requirements, minimizing delays in their appointments and/or medical treatments. These methodologies are compared with the current mode of assignment used. To test the previously proposed solutions, data provided by the Arturo López Perez Foundation (FALP) will be used.

关键词： Schedules scheduling algorithms Stochastic processes Medical treatment Programming Oncology Dynamic scheduling Prediction algorithms Mixed integer linear programming Predictive analytics

来源：评论

学校读者我要写书评

暂无评论

Dynamic scheduling Using Multi-scale Graph Attention Neural Network

Dynamic Scheduling Using Multi-scale Graph Attention Neural ...

引用

International Conference on Power and Energy Technology (ICPET)

作者： Hua-Kun Que Sheng Huang Xiao Jiang Rui-Biao Xie Qian-Qian Cai Guangdong Power Grid Corporation Metrology Center Guangzhou China

ISBN: (数字)9798350375855

ISBN: (纸本)9798350375862

Dynamic scheduling issues have garnered considerable attention in the realm of network topology-based research. Within power systems, efficient resource allocation and link scheduling, rooted in system networks, are paramount for optimizing power resource utilization and operational efficacy. This paper delves into the distributed link scheduling task within system networks, where a prevalent approach relies on Maximum Weighted Independent Set (MWIS) and greedy algorithms due to the versatility and broad applicability. Nonetheless, these greedy methodologies tend to disregard different ranges of topological information within networks. Different business and regional conflict network structures in power system vary, and different neighborhoods in the conflict network contribute differently to the node's unity function. To adapt to different network structures and comprehensively consider the roles of nodes within different ranges of neighborhoods, this paper proposes a link scheduling algorithm based on multi-scale graph attention feature embedding i.e. Multi-scale Graph Attention based MWIS solver. This solver integrates the combined effects of nodes within multiple neighborhood ranges in the conflict graph by defining different attention scales for graph feature embedding, thereby improving the unity function of the generated scheduling strategy. Experimental results validate the superiority of the proposed multi-scale graph attention feature embedding-based link scheduling algorithm over conventional approaches, affirming the efficacy of multi-scale graph attention features.

关键词： Greedy algorithms scheduling algorithms Heuristic algorithms Dynamic scheduling Graph neural networks Power systems Resource management Business

来源：评论

学校读者我要写书评

暂无评论

Low Earth Orbit Satellite scheduling Optimization Based on Deep Reinforcement Learning

Low Earth Orbit Satellite Scheduling Optimization Based on D...

引用

International Conference on Information and Communication Technology Convergence (ICTC)

作者： Junyoung Kim Soyi Jung Dept. Artificial Intelligence Convergence Network Ajou University Suwon South Korea Dept. Electrical and Computer Engineering Ajou University Suwon South Korea

ISBN: (数字)9798350364637

ISBN: (纸本)9798350364644

In next-generation 6G scenarios, non-terrestrial net-works employing low Earth orbit (LEO) satellites will be pivotal in achieving ultra-wide coverage, ultra-connectivity, and ultra-precision. Although LEO satellites provide comprehensive global coverage, their rapid mobility introduces frequent handovers, requiring sophisticated scheduling to maintain uninterrupted service. This paper proposes a deep reinforcement learning-based scheduling algorithm in order to improve service rate and continuity for terrestrial users in multi-LEO) environments.

关键词： Satellites scheduling algorithms Low earth orbit satellites Handover Deep reinforcement learning Timing Information and communication technology Optimization Next generation networking Convergence

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：