检索结果-内蒙古大学图书馆

computation scheduling for Distributed Machine Learning With Straggling Workers

IEEE TRANSACTIONS ON SIGNAL PROCESSING 2019年第24期67卷 6270-6284页

作者： Amiri, Mohammad Mohammadi Gunduz, Deniz Princeton Univ Dept Elect Engn Princeton NJ 08544 USA Imperial Coll London Dept Elect & Elect Engn London SW7 2AZ England

We study scheduling of computation tasks across $n$ workers in a large scale distributed learning problem with the help of a master. computation and communication delays are assumed to be random, and redundant computations are assigned to workers in order to tolerate stragglers. We consider sequential computation of tasks assigned to a worker, while the result of each computation is sent to the master right after its completion. Each computation round, which can model an iteration of the stochastic gradient descent (SGD) algorithm, is completed once the master receives $k$ distinct computations, referred to as the computation target. Our goal is to characterize the average completion time as a function of the computation load, which denotes the portion of the dataset available at each worker, and the computation target. We propose two computation scheduling schemes that specify the tasks assigned to each worker, as well as their computation schedule, i.e., the order of execution. Assuming a general statistical model for computation and communication delays, we derive the average completion time of the proposed schemes. We also establish a lower bound on the minimum average completion time by assuming prior knowledge of the random delays. Experimental results carried out on Amazon EC2 cluster show a significant reduction in the average completion time over existing coded and uncoded computing schemes. It is also shown numerically that the gap between the proposed scheme and the lower bound is relatively small, confirming the efficiency of the proposed scheduling design.

关键词： Task analysis Delays Processor scheduling computational modeling Encoding Schedules Decoding Distributed machine learning uncoded computing computation scheduling straggling workers

来源：评论

学校读者我要写书评

暂无评论

DECO: Joint computation scheduling, Caching, and Communication in Data-Intensive Computing Networks

引用

IEEE-ACM TRANSACTIONS ON NETWORKING 2022年第3期30卷 1058-1072页

作者： Kamran, Khashayar Yeh, Edmund Ma, Qian Pinterest Inc San Francisco CA 94107 USA Northeastern Univ Dept Elect & Comp Engn Boston MA 02115 USA Sun Yat Sen Univ Sch Intelligent Syst Engn Guangzhou 510275 Peoples R China

Driven by technologies such as IoT-enabled health care, machine learning applications at the edge, and industrial automation, mobile edge and fog computing paradigms have reinforced a general trend toward decentralized computing, where any network node can route traffic, compute tasks, and store data, possibly at the same time. In many such computing environments, there is a need to cache significant amounts of data, which may include large data sets, machine learning models, or executable code. In this work, we propose a framework for joint computation scheduling, caching, and request forwarding within such decentralized computing environments. We first characterize the stability region of a ``genie-aided'' computing network where data required by computation are instantly accessible, and develop a throughput optimal control policy for this model. Based on this, we develop a practically implementable distributed and adaptive algorithm, and show that it exhibits superior performance in terms of average task completion time, when compared to several baseline policies.

关键词： Task analysis Processor scheduling Throughput computational modeling Edge computing Network topology Low latency communication Edge computing fog computing computation scheduling distributed computing data-intensive computing caching

来源：评论

学校读者我要写书评

暂无评论

Dynamic computation scheduling for hybrid energy mobile edge computing networks

引用

JOURNAL OF SYSTEMS ARCHITECTURE 2024年 155卷

作者： Bi, Ran Sun, Liang Sun, Yiwei Han, Meng Deng, Qingxu Northeastern Univ Qinhuangdao Sch Comp & Commun Engn Qinhuangdao Peoples R China Hebei Key Lab Marine Percept Network & Data Proc Beijing Peoples R China Dalian Univ Technol Sch Comp Sci & Technol Dalian Peoples R China Zhejiang Univ Binjiang Res Inst Hangzhou Peoples R China Northeastern Univ Sch Comp Sci & Engn Shenyang Peoples R China

The rapid expansion of Mobile Edge Computing (MEC) is driven by the growing demand for resource-intensive applications within the Internet of Things. computation offloading allows these applications to be executed at the network edge, but it leads to significant electricity expenses for network operators. To mitigate these costs, one promising solution is to power base stations (BSs) with a hybrid energy supply that combines unpredictable harvested energy with stable energy from the smart grid. This paper investigates joint computation scheduling for mobile devices (MDs) and resource allocation in a MEC network incorporating hybrid energy sources. Our objective is to maximize long-term time-averaged service utility by optimizing parameters such as BS battery supply, harvestable energy, CPU frequency, transmission power, task-partition factor, and MD-BS associations. To tackle this complex problem, we exploit the Lyapunov optimization framework to decompose it into deterministic subproblems for each time slot and propose an online network service utility maximization scheduling (NSUMS) algorithm. Experimental results show that our algorithm outperforms benchmark schemes in service utility and energy expenditure, improving the completion ratio by 32%, reducing the failure rate by 80%, and decreasing MD energy consumption by 28%.

关键词： Mobile edge computing computation scheduling Energy harvesting Task offloading

来源：评论

学校读者我要写书评

暂无评论

Task offloading and computational scheduling in RIS-assisted low Earth orbit satellite communication networks

引用

VEHICULAR COMMUNICATIONS 2025年 53卷

作者： Wang, Yin Gui, Kang'An Chongqing Univ Posts & Telecommun Sch Commun & Informat Engn 2 Chongwen Rd Chongqing 400065 Peoples R China

This article investigates the joint optimization of task offloading and computation scheduling in low Earth orbit (LEO) satellite networks, where each LEO satellite is equipped with a reconfigurable intelligent surface (RIS). By considering the inherent characteristics of tasks and the energy consumption associated with task execution, we define a system utility function and formulate the problem as a constrained utility maximization problem. To address this optimization challenge, we first propose a priority-based task offloading and computation scheduling strategy tailored for single-satellite execution scenarios. Subsequently, we extend this approach to multi-satellite collaborative task execution scenarios, where a knapsack algorithm-based strategy is developed to optimize task allocation and scheduling. To underscore the advantages of the proposed RIS-assisted multi-satellite framework, we introduce a comparative analysis with a non-RIS-assisted multi-satellite offloading mode. Extensive simulations conducted in Satellite Tool Kit (STK) and MATLAB demonstrate that the RIS-assisted multi-satellite mode significantly outperforms its non-RIS counterpart in terms of system utility and energy efficiency. The results validate the effectiveness of the proposed algorithms and highlight the potential of RIS technology in enhancing the performance of LEO satellite networks.

关键词： LEO satellites RIS Task offloading computation scheduling

来源：评论

学校读者我要写书评

暂无评论

Accelerating End-to-End Deep Learning Workflow With Codesign of Data Preprocessing and scheduling

引用

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 2021年第7期32卷 1802-1814页

作者： Cheng, Yang Li, Dan Guo, Zhiyuan Jiang, Binyao Geng, Jinkun Bai, Wei Wu, Jianping Xiong, Yongqiang Tsinghua Univ Beijing 100084 Peoples R China Beihang Univ Beijing 100191 Peoples R China Microsoft Res Beijing 100080 Peoples R China Shanghai Jiao Tong Univ Shanghai 200240 Peoples R China

In this article, we investigate the performance bottleneck of existing deep learning (DL) systems and propose DLBooster to improve the running efficiency of deploying DL applications on GPU clusters. At its core, DLBooster leverages two-level optimizations to boost the end-to-end DL workflow. On the one hand, DLBooster selectively offloads some key decoding workloads to FPGAs to provide high-performance online data preprocessing services to the computing engine. On the other hand, DLBooster reorganizes the computational workloads of training neural networks with the backpropagation algorithm and schedules them according to their dependencies to improve the utilization of GPUs at runtime. Based on our experiments, we demonstrate that compared with baselines, DLBooster can improve the image processing throughput by 1.4x - 2.5x and reduce the processing latency by 1/3 in several real-world DL applications and datasets. Moreover, DLBooster consumes less than 1 CPU core to manage FPGA devices at runtime, which is at least 90 percent less than the baselines in some cases. DLBooster shows its potential to accelerate DL workflows in the cloud.

关键词： Data preprocessing Training Field programmable gate arrays Task analysis Graphics processing units Artificial neural networks Hardware Deep learning data preprocessing workload offloading computation scheduling FPGAs

来源：评论

学校读者我要写书评

暂无评论

A System Utility Optimization-based Task Offloading and Computing scheduling Algorithm for Low Earth Orbit Satellite Networks 59

A System Utility Optimization-based Task Offloading and Comp...

引用

59th Annual IEEE International Conference on Communications (IEEE ICC)

作者： Wang, Yin Gui, Kang'An Chai, Rong Chongqing Univ Posts & Telecommun Sch Commun & Informat Engn Chongqing Peoples R China

ISBN: (纸本)9781728190549

In this paper, we investigate the joint task offloading and computing scheduling in low earth orbit (LEO) satellite networks, which consists of a number of ground terminals (GTs) and multiple LEO satellites. Considering the characteristics of tasks and the energy consumption of task execution, we define a system utility function, and formulate the problem of task offloading and computing scheduling as a constrained system utility function maximization problem. The formulated optimization problem is a mixed integer non-linear programming problem which cannot be solved conveniently. To solve the formulated optimization problem, we first propose local computing scheduling algorithm to determine the task local partition strategy. And then priority-based offloading and computing scheduling algorithm is proposed to determine the task offloading strategy. The effectiveness of the proposed algorithm is verified through simulations in satellite tool kit (STK) and MATLAB.

关键词： system utility low earth orbit satellite mobile edge computing task offloading computation scheduling

来源：评论

学校读者我要写书评

暂无评论

ResiRCA: A resilient energy harvesting ReRAM crossbar-based accelerator for intelligent embedded processors 26

ResiRCA: A resilient energy harvesting ReRAM crossbar-based ...

引用

26th IEEE International Symposium on High Performance Computer Architecture (HPCA)

作者： Qiu, Keni Jao, Nicholas Zhao, Mengying Mishra, Cyan Subhra Gudukbay, Gulsum Jose, Sethu Sampson, Jack Kandemir, Mahmut Taylan Narayanan, Vijaykrishnan Capital Normal Univ Beijing Peoples R China Shandong Univ Qingdao Peoples R China Penn State Univ University Pk PA 16802 USA

ISBN: (数字)9781728161495

ISBN: (纸本)9781728161495

Many recent works have shown substantial efficiency boosts from performing inference tasks on Internet of Things (IoT) nodes rather than merely transmitting raw sensor data. However, such tasks, e.g., convolutional neural networks (CNNs), are very compute intensive. They are therefore challenging to complete at sensing-matched latencies in ultra-low-power and energy-harvesting IoT nodes. ReRAM crossbar-based accelerators (RCAs) are an ideal candidate to perform the dominant multiplication-and-accumulation (MAC) operations in CNNs efficiently, but conventional, performance-oriented RCAs, while energy-efficient, are power hungry and ill-optimized for the intermittent and unstable power supply of energy-harvesting IoT nodes. This paper presents the ResiRCA architecture that integrates a new, lightweight, and configurable RCA suitable for energy harvesting environments as an opportunistically executing augmentation to a baseline sense-and-transmit battery-powered IoT node. To maximize ResiRCA throughput under different power levels, we develop the ResiSchedule approach for dynamic RCA reconfiguration. The proposed approach uses loop tiling-based computation decomposition, model duplication within the RCA, and inter-layer pipelining to reduce RCA activation thresholds and more closely track execution costs with dynamic power income. Experimental results show that ResiRCA together with ResiSchedule achieve average speedups and energy efficiency improvements of 8x and 14 x respectively compared to a baseline RCA with intermittency-unaware scheduling.

关键词： Energy harvesting ReRAM crossbar CNN Reconfigurable hardware Loop tiling computation scheduling

来源：评论

学校读者我要写书评

暂无评论

Joint allocation of computation and communication resources in multiuser mobile cloud computing

Joint allocation of computation and communication resources ...

引用

IEEE 14th Workshop on Signal Processing Advances in Wireless Communications (SPAWC)

作者： Barbarossa, Sergio Sardellitti, Stefania Di Lorenzo, Paolo Univ Roma La Sapienza Dept Informat Eng Elect & Telecommun I-00184 Rome Italy

ISBN: (纸本)9781467355773

Mobile cloud computing is offering a very powerful storage and computational facility to enhance the capabilities of resource-constrained mobile handsets. However, full exploitation of the cloud computing capabilities can be achieved only if the allocation of radio and computational capabilities is performed jointly. In this paper, we propose a method to jointly optimize the transmit power, the number of bits per symbol and the CPU cycles assigned to each application in order to minimize the power consumption at the mobile side, under an average latency constraint dictated by the application requirements. We consider the case of a set of mobile handsets served by a single cloud and we show that the optimization leads to a one-to-one relationship between the transmit power and the percentage of CPU cycles assigned to each user. Based on our optimization, we propose then a computation scheduling technique and verify the stability of the computations queue. Then we show how these queues are affected by the degrees of freedom of the channels between mobile handsets and server.

关键词： Mobile cloud computing radio resource allocation computation scheduling

来源：评论

学校读者我要写书评

暂无评论

Uantifying data locality in dynamic parallelism in GPUs

Uantifying data locality in dynamic parallelism in GPUs

引用

Proceedings of the ACM on Measurement and Analysis of Computing Systems

作者： Tang, Xulong Pattnaik, Ashutosh Kayiran, Onur Jog, Adwait Kandemir, Mahmut Taylan Das, Chita Pennsylvania State University State CollegePA16801 United States College of William and Mary WilliamsburgVA23185 United States Advanced Micro Devices Inc Santa ClaraCA95054 United States

GPUs are becoming prevalent in various domains of computing and are widely used for streaming (regular) applications. However, they are highly ineicient when executing irregular applications with unstructured inputs due to load imbalance. Dynamic parallelism (DP) is a new feature of emerging GPUs that allows new kernels to be generated and scheduled from the device-side (GPU) without the host-side (CPU) intervention to increase parallelism. To eiciently support DP, one of the major challenges is to saturate the GPU processing elements and provide them with the required data in a timely fashion. There have been considerable eforts focusing on exploiting data locality in GPUs. However, there is a lack of quantitative analysis of how irregular applications using dynamic parallelism behave in terms of data reuse. In this paper, we quantitatively analyze the data reuse of dynamic applications in three diferent granularities of schedulable units: kernel, work-group, and wavefront. We observe that, for DP applications, data reuse is highly irregular and is heavily dependent on the application and its input. Thus, existing techniques cannot exploit data reuse efectively for DP applications. To this end, we irst conduct a limit study on the performance improvements that can be achieved by hardware schedulers that are provided with accurate data reuse information. This limit study shows that, on an average, the performance improves by 19.4% over the baseline scheduler. Based on the key observations from the quantitative analysis of our DP applications, we next propose LASER, a Locality-Aware SchedulER, where the hardware schedulers employ data reuse monitors to help make scheduling decisions to improve data locality at runtime. Our experimental results on 16 benchmarks show that LASER, on an average, can improve performance by 11.3%. © 2018 Association for Computing Machinery.

关键词： Benchmarking Computer graphics Computer hardware Graphics processing unit Hardware Program processors scheduling Cache locality computation scheduling Data reuse Dynamic applications Irregular applications Performance evaluations Performance improvements scheduling decisions

来源：评论

学校读者我要写书评

暂无评论

A survey on edge computing for wearable technology

引用

DIGITAL SIGNAL PROCESSING 2022年 125卷

作者： Jin, Xinqi Li, Lingkun Dang, Fan Chen, Xinlei Liu, Yunhao Tsinghua Univ Beijing Peoples R China Michigan State Univ E Lansing MI USA

Smart wearable devices have become more and more popular in our daily life due to their unique power of "wearing-while-using. " However, requirements of light weight and compact size lead to limited on device resources in most wearable products, which hamper the development of wearable technology. Edge computing provides an opportunity for wearable devices to access more resources without violating the constraints on weight and size. In this article, we first investigate the drawbacks of wearable devices and explore the potential of addressing such drawbacks by edge computing. Then we conduct a comprehensive survey on existing works from four aspects, i.e., computation scheduling, information perception, energy-saving, and security. Finally, we point out several future research directions worth our attention. We believe that wearable devices enhanced with edge computing technologies would bring more benefits and convenience to our life shortly. (C) 2021 Elsevier Inc. All rights reserved.

关键词： Wearable technology Edge computing computation scheduling Energy-saving

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：