检索结果-内蒙古大学图书馆

Key Flow First Prioritized Flow scheduling Strategy in Multi-Tenant Data Centers

IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT 2024年第3期21卷 3264-3277页

作者： Tao, Xudong Qian, Xiaoyan Han, Lei Fan, Weibei Shi, Yuzhou Zhu, Xinrui Li, Zhiyu Wei, Shuwen Xu, Rui Nanjing Univ Aeronaut & Astronaut Coll Civil Aviat Nanjing 211106 Peoples R China Nanjing Univ Posts & Telecommun Sch Comp Nanjing 210023 Peoples R China Nanjing Univ Aeronaut & Astronaut Nanjing 210016 Peoples R China

The mixed flow in multi-tenant data centers presents a challenge for priority flow scheduling due to the coexistence of various requirements such as latency and throughput. To address this issue, we propose Key Flow First (KFF), a balanced scheduling algorithm suitable for mixed flows in multi-tenant data centers. Firstly, KFF categorizes flows into Latency-Sensitive Flows (LS Flow) and Throughput-Demanding Flows (TD Flow) based on the Quality of Service (QoS) of their application sources. Secondly, it further differentiates flows into Mice Flows and Elephants Flows based on the amount of already sent bytes. Thirdly, KFF employs the Multi-Level Feedback Queue (MLFQ) threshold update algorithm and a priority-based strict forwarding mechanism. By avoiding reliance on complex flow priors, KFF consistently maintains reasonable scheduling of mixed flows under different load scenarios. Experimental results demonstrate that KFF effectively reduces the real-time load on the network and achieves good performance in terms of MAX (Shortest Job First (SJF), Earliest Deadline First (EDF)) performance under diverse load conditions. Compared to PIAS, KFF reduces the FCT slow down of deadline flows by nearly 60% under high TD loads;compared to Karuma and Time Deadline Aware pFabric (TDA-pFabric), KFF reduces the flow completion time (FCT) slow down of non-deadline Mice flows by over 90% under high LS loads and meanwhile guaranteeing nearly 0 deadline miss rate.

关键词： Computer networks Data centers scheduling algorithms Heuristic algorithms Communication system traffic scheduling Quality of service Data center networks traffic engineering flow scheduling fair queueing

来源：评论

学校读者我要写书评

暂无评论

RIS-Assisted Transmission scheduling for Train-to-Train Communication System

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第10期73卷 15020-15033页

作者： Li, Wenqing Ai, Bo Niu, Yong Han, Zhu Wang, Ning Zhong, Zhangdui He, Ruisi Beijing Jiaotong Univ Sch Elect & Informat Engn Beijing 100044 Peoples R China Univ Houston Dept Elect & Comp Engn Houston TX 77004 USA Kyung Hee Univ Dept Comp Sci & Engn Seoul 446701 South Korea Zhengzhou Univ Sch Informat Engn Zhengzhou 450001 Peoples R China

Millimeter wave (mmWave) has been considered as an enabler to alleviate the spectrum shortage problem while providing high data rates in high speed railway (HSR) communications. However, a mmWave channel has a larger path loss, and therefore is sensitive to blockage. In such a high speed mobile environment with fast-changing channels, considering the characteristics of mmWave, how to achieve reliable transmission scheduling in a train-to-train communication has become a critical issue, especially from the safety point of view. In this paper, we investigate a mmWave train-to-train communication system with the fixed frame structure, where two trains communicate with each other directly by mobile relays (MRs) deployed on top of cabin roofs. Due to the limited communication distance by MRs and the nature that mmWave is prone to blockage, reconfigurable intelligent surface (RIS) is placed between two trains to rebuild line of sight (LoS) link when original link is blocked or beyond the distance range. We aim to maximize the number of completely scheduled flows with different quality of service (QoS) requirements via optimizing the discrete phase shifts of RIS as well as flow scheduling. In particular, we consider multiple-frame scheduling to make use of the predictability of trains' trajectory. To solve this nonlinear integer programming problem, we decompose the joint scheduling problem into two subproblems. We first optimize the discrete phase shifts of RIS, and then build an interference-based graph of flows while transforming the transmission constraints into interference relationships. The goal of the problem can be realized through the proposed scheduling algorithm. Finally, simulation results show that the proposed algorithm outperforms three baseline schemes in the literature in terms of the number of the completely scheduled flows and system throughput.

关键词： Millimeter wave communication Reconfigurable intelligent surfaces Rail transportation Quality of service scheduling algorithms Optimization Job shop scheduling MmWave high speed railway (HSR) reconfigurable intelligent surface (RIS) train-to-train communication

来源：评论

学校读者我要写书评

暂无评论

scheduling Coflows in Hybrid Optical-Circuit and Electrical-Packet Switches With Performance Guarantee

引用

IEEE-ACM TRANSACTIONS ON NETWORKING 2024年第3期32卷 2299-2314页

作者： Wang, Xin Shen, Hong Tian, Hui Sun Yat Sen Univ Sch Comp Sci & Engn Guangzhou 510275 Peoples R China Cent Queensland Univ Sch Engn & Technol Brisbane Qld 4000 Australia Griffith Univ Sch Informat & Commun Technol Nathan Qld 4111 Australia

scheduling of coflows, each a collection of parallel flows sharing the same objective, is an important task of data transmission that arises in the networks supporting data-intensive applications such as data center networks (DCNs). The hybrid switch design combining the optical circuit switch (OCS) and electrical packet switch (EPS) for transmitting high-volume and low-volume traffic separately has received considerable research attention. To support this design, efficient scheduling of coflows on hybrid network links is crucial for reducing the overall communication time. However, because it needs to consider both reconfiguration delay of circuit switching in the OCS and bandwidth limitation of packet switching in the EPS, coflow scheduling on hybrid network links is more challenging than on monotonic network links of either OCS or EPS. The existing coflow scheduling algorithms in hybrid switches are all heuristic and provide no performance guarantees. In this work, we first propose an approximation algorithm with a worst-case performance guarantee of 2 tau , where tau <= N is the maximum number of non-zero elements of each row and column of coflow's demand matrix, for single coflow scheduling in an NxN hybrid switch to minimize the coflow completion time (CCT). We then extend the algorithm for scheduling multiple coflows to minimize the total weighted CCT with a provable performance guarantee of mu tau(max) , where mu=4M & sdot;w(max)/w(min) , tau(max)=max(1) <= m <= M tau m <= N , w(max) and w(min) are respectively the maximum and minimum weights of the M coflows. Extensive simulations using Facebook data traces show that our algorithms outperform the state-of-the-art coflow scheduling schemes. Specifically, our algorithms transmit a single coflow up to 1.08 x faster than Solstice (hybrid switch) and 1.42 x faster than Reco-Sin (pure OCS), and multiple coflows u

关键词： Optical switches scheduling Approximation algorithms Optical packet switching scheduling algorithms Delays Bandwidth Coflow scheduling optical circuit switch electrical packet switch datacenter network approximation algorithm

来源：评论

学校读者我要写书评

暂无评论

IRFS: A CQF scheduling Method Integrating Queue Resources and Flow Features in Time-Sensitive Networking

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第10期73卷 14201-14211页

作者： Sun, Wenjing Zou, Yuan Guan, Nan Zhang, Xudong Fan, Jie Meng, Yihao Beijing Inst Technol Collaborat Innovat Ctr Elect Vehicles Beijing Sch Mech Engn Natl Engn Lab Elect Vehicles Beijing 100081 Peoples R China City Univ Hong Kong Dept Comp Sci Hong Kong Peoples R China

Time-sensitive networking (TSN) has revolutionized Ethernet with real-time and deterministic transmission capabilities, making it one of the most potential solutions for future vehicular and industrial networks. Compared to time-aware shaper (TAS), the cyclic queuing and forwarding (CQF) protocol simplifies the gate control list (GCL) configuration process, reducing the deployment difficulty of TSN in large-scale networks. Much research has proposed incremental scheduling approaches for the CQF. However, existing methods often inadequately consider and insufficiently integrate network and flow characteristics, limiting scheduling performance. This paper introduces a novel CQF scheduling method, IRFS, which integrates queue resources and flow features for efficient searching of scheduling priority, routing path, and start offset. A priority sorting function is proposed that deeply combines network and flow characteristics while considering both spatial and temporal resource allocation. IRFS achieves efficient scheduling and load balancing by constructing combinations of $(flow, path, offset)$, where the elements respectively represent flow features, the spatial distribution, and the temporal distribution of resources. The IRFS is validated in different network scenarios, including simple, complex, and In-Vehicle Networking (IVN) settings. It is compared against other state-of-the-art CQF scheduling algorithm. The IRFS demonstrates superior performance in scheduling success rate, load balancing, and computation time across these scenarios.

关键词： Job shop scheduling Sorting Logic gates Routing Switches scheduling algorithms Resource management Time-sensitive networking cyclic queuing and forwarding traffic scheduling resource mapping

来源：评论

学校读者我要写书评

暂无评论

An evaluation of scheduling algorithms in LTE based 4G networks

An evaluation of scheduling algorithms in LTE based 4G netwo...

引用

International Conference on Emerging Technologies, ICET

作者： Qurat-ul-Ain Syed Riaz ul Hassnain Mudassir Shah Sahibzada Ali Mahmud Department of Electrical Engineering University of Engineering and Technology Peshawar Pakistan Department of Electronics Engineering University of Engineering and Technology Peshawar Abbottabad Campus Pakistan Department of Electrical Engineering Sarhad University of Science and IT Peshawar Pakistan

The intent of 4G technology is to converge high speed data application requirements and compete with other technologies. Different techniques i.e relaying, carrier aggregation, multiple input multiple output, and heterogeneous networks provide higher throughputs, low latency and enable LTE to become a Standard for Wireless broadband. Due to almost exponential increase in demand for high data rates it is expected that network will be occupied by more resource demanding applications. Therefore efficient scheduling of radio resources is required for better performance of LTE system. Proportional Fair scheduler is adopted to LTE system to achieve Orthogonal Frequency Division Multiple Access (OFDMA) potentiality in Time Domain systems (TDS) and Frequency Domain scheduling systems (FDS). This paper aims at enhancing the functionality of Proportional Fair algorithm by making future estimation of the channel quality index (CQI) to offer higher data rate among all users according to their running applications, at high mobility. Performance of proposed algorithm on the downlink is measured in terms of throughput and block error rate using a MATLAB-based system level simulator.

关键词： Throughput Long Term Evolution scheduling algorithms Indexes Resource management MIMO OFDM

来源：评论

学校读者我要写书评

暂无评论

On fairness in LTE downlink MAC scheduling algorithms

On fairness in LTE downlink MAC scheduling algorithms

引用

International Conference on Information and Communication Technology Research (ICTRC)

作者： Ismael I. Al-Shiab Rana E. Ahmed Dept. of Computer Science and Engineering American University of Sharjah Sharjah UAE

A downlink MAC scheduler aims to improve the radio resources utilization in the Long Term Evolution (LTE) network. The MAC scheduler is responsible for sharing the available radio Resource Blocks (RBs) among different User Equipment (UEs) in time and frequency domains. Several downlink MAC schedulers have been proposed and implemented; however, providing fairness in terms of granting resources to different UEs with different QoS requirements remains a challenge. This paper presents the results of a comparative simulation study with respect to a fairness criterion for three well-known downlink schedulers.

关键词： Throughput Downlink Long Term Evolution Indexes Measurement Aggregates scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Decentralized scheduling for Concurrent Tasks in Mobile Edge Computing via Deep Reinforcement Learning

引用

IEEE TRANSACTIONS ON MOBILE COMPUTING 2024年第4期23卷 2765-2779页

作者： Fan, Ye Ge, Jidong Zhang, Sheng Wu, Jie Luo, Bin Nanjing Univ State Key Lab Novel Software Technol Nanjing 210008 Jiangsu Peoples R China Nanjing Univ Software Inst Nanjing 210008 Jiangsu Peoples R China Temple Univ Ctr Networked Comp Philadelphia PA 19122 USA

Mobile Edge Computing (MEC) is a promising solution to enhance the computing capability of resource-limited networks. A fundamental problem in MEC is efficiently offloading tasks from user devices to edge servers. However, there still exists a gap to deploy in real-world environments: 1) traditional centralized approaches needs complete information of edge network, ignoring the communication costs generated by synchronization, 2) previous works do not consider concurrent computation on edge servers, which may cause dynamic changes in the environment, and 3) the scheduling algorithm should deliver individualized decisions for different users independently and with high efficiency To solve this mismatch, we studied a multi-user task offloading problem where user devices make offloading decisions independently. We consider the concurrent execution of tasks and formulate a non-divisible and delay-aware task offloading problem to jointly minimize the dropped task ratio and long-term latency. We propose a decentralized task scheduling algorithm based on DRL that makes offloading decisions without knowing the information of other user devices. We employ Double-DQN, Dueling-DQN, Prioritized Replay Memory, and Recurrent Neural Network (RNN) techniques to improve the algorithm's performance. The results of simulation experiments show that our method can significantly reduce the long-term latency and dropped task ratio compared to the baseline algorithms.

关键词： Task analysis Servers Costs Heuristic algorithms Training scheduling algorithms Deep learning Deep reinforcement learning deep q-learning mobile edge computing resource allocation task offloading

来源：评论

学校读者我要写书评

暂无评论

Optimal planning and scheduling of information processes during interaction among mobile objects

引用

INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 2024年第16期62卷 5905-5924页

作者： Boris, Sokolov Vitaly, Ushakov Valerii, Zakharov Russian Acad Sci St Petersburg Fed Res Ctr 39 14th Line VO St Petersburg 199178 Russia St Petersburg State Univ Aerosp Instrumentat St Petersburg Russia

The purpose of the research was to improve the quality of information process control in the interaction of mobile object grouping. This proposal is based on developing and implementing new combined models with algorithms for optimal planning and scheduling of information processes during interaction among mobile objects. The proposed approach would guarantee the required level of aircraft maintenance at the airport. The novel aspect of the proposed solution is that it divides the spatially distributed task of selecting a plan and a schedule for information processes in mobile objects into two subtasks, describing the spatial and temporal features of the original task. The spatial component of the original problem is described as a static model. Using this model, the optimal distribution of information operations over the resources of mobile objects is carried out without binding these operations to time. The time component is described as a dynamic model, within which the information operations are tied to time and the corresponding resources. Examples of construction and research of synthesized plans and schedules are ***: DM, Dynamic model;FIFO, First-In First-Out;Gr, Grouping;HSC, Hardware and software complex;ITS, Intelligent Transport System;IP, Information Processes;MO, Mobile object;PC, Programmed control;SM, Static model;SMS, Special mathematic support;LDM, Logical-dynamic models

关键词： static planning dynamic scheduling scheduling algorithms optimal program control sequential decomposition intelligent transport systems

来源：评论

学校读者我要写书评

暂无评论

DB-ACO: A Deadline-Budget Constrained Ant Colony Optimization for Workflow scheduling in Clouds

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2024年第2期21卷 1564-1579页

作者： Tao, Siyuan Xia, Yuanqing Ye, Lingjuan Yan, Ce Gao, Runze Beijing Inst Technol Sch Math & Stat Beijing 100081 Peoples R China Beijing Inst Technol Sch Automat Beijing 100081 Peoples R China

With the development of cloud computing, a growing number of workflows are deployed in cloud platform that can dynamically provide cloud resources on demand for users. In clouds, one basic problem is how to schedule workflow under the deadline constraint and minimize the execution cost. As the capability of cloud resources getting higher, the required cost is also rising. Capability of some resources exceeds the need of users, which leads to higher cost, and the budget of users should be considered. In this paper, a novel scheduling algorithm, named DB-ACO, is proposed to minimize the execution cost for the workflow with deadline and budget constraints. DB-ACO is verified on four typical scientific workflows, and the experiments results show it outperforms four state-of-the-art methods, especially for *** to Practitioners-Budget and deadline are important requirements for users in cloud computing, which are used as constraints. Extensive works have been devoted to minimize the cost of workflows execution with different scheduling strategies. However, most of them only consider one single constraint and assume the constraint is simple and loose, which is impractical in actual scenarios due to higher requirement of users. This paper investigates a novel scheduling algorithm DB-ACO to optimize cost under budget and deadline. DB-ACO combines heuristic and meta-heuristic, it uses ant colony optimization to optimize the execution cost under the deadline and budget constraints: each ant sorts tasks on the basis of the combination of the pheromone trail and heuristic information, the deadline and budget are distributed fairly to each task by a novel distribution method, then the service selection rules are introduced to build solution.

关键词： Task analysis Costs Cloud computing scheduling Metaheuristics scheduling algorithms Schedules ant colony optimization deadline budget workflow scheduling

来源：评论

学校读者我要写书评

暂无评论

Cloud Computing Resource scheduling Algorithm Based on Unsampled Collaborative Knowledge Graph Network

引用

IEEE ACCESS 2024年 12卷 186476-186483页

作者： Sun, Haichuan Gu, Liang Dong, Chenni Ma, Xin Liu, Zeyu Li, Zhenxi State Grid Shanxi Elect Power Co Informat & Commun Branch Taiyuan 030021 Peoples R China Beijing CLP Puhua Informat Technol Co Ltd Beijing 100089 Peoples R China

A cloud computing resource scheduling algorithm based on sampled collaborative knowledge graph network is designed to address the issues of lag in the process of cloud computing resource scheduling, high overall load rate, and large transient amplitude and phase errors. Based on graph convolutional neural networks, analyze the target load of cloud platforms, construct multi hop data transmission paths one by one, and perform deep level information load balancing;Establish a multiplexing information transmission model, correct the initial weights of graph convolutional neural networks, combine reverse transmission calculation methods, integrate and balance cloud computing resources, and confirm the optimal resource scheduling plan;Integrating class convolution and human-machine interaction attention mechanism, the value of the previous time series neural unit is transferred to the current neural unit, and the classification output sequence of knowledge graph relational data feature fragments is analyzed. The knowledge graph data fragments are processed based on class convolution and human-machine interaction attention mechanism, and different sizes of linear aggregators are used to capture deep level information, completing the design of cloud computing resource scheduling algorithm. The experimental results show that although the load rate is on the rise, the highest is only 89%, and the scheduling rate is relatively high, ranging from 38.9 to 43.1bps;The energy consumption is relatively low, not exceeding 40.106mW. In terms of transient amplitude and phase, the proposed method can control the error within 2.0. Ensure the efficiency and practical application effectiveness of cloud computing resource scheduling algorithms.

关键词： Cloud computing Knowledge graphs Transient analysis Resource management Convolutional neural networks scheduling algorithms Load modeling Relays Multiplexing Graph convolutional networks Knowledge graph graph convolutional neural network cloud computing resource scheduling

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：