基于SAC的多服务移动边缘计算中任务卸载和资源配置算法

多服务移动边缘计算网络环境中的不同服务的缓存要求、受欢迎程度、计算要求以及从用户传输到边缘服务器的数据量是随时间变化的。如何在资源有限的边缘服务器中调整总服务类型的缓存子集,并确定任务卸载目的地和资源分配决策,以获得最佳的系统整体性能是一个具有挑战性的问题。为了解决这一难题,首先将优化问题转换为马尔可夫决策过程,然后提出了一种基于软演员—评论家（soft actor-critic,SAC）的深度强化学习算法来同时确定服务缓存和任务卸载的离散决策以及上下带宽和计算资源的连续分配决策。算法采用了将多个连续动作输出转换为离散的动作选择的有效技巧,以应对连续—离散混合行动空间所带来的关键设计挑战,提高算法决策的准确性。此外,算法集成了一个高效的奖励函数,增加辅助奖励项来提高资源利用率。广泛的数值结果表明,与其他基线算法相比,提出的算法在有地减少任务的长期平均完成延迟的同时也具有良好的稳定性。     

车载边缘计算中基于信誉值的计算卸载方法研究*

将移动边缘计算（Mobile edge computing, MEC）引入车载自组网形成车载边缘计算,从而使服务提供商直接利用MEC服务器在网络边缘服务用户,以提升用户体验质量和丰富用户满意度。随后,研究在车载边缘计算环境下车辆用户的计算卸载问题。针对此问题,提出相应的系统模型与使用讨价还价博弈方法以解决MEC服务器如何根据不同的任务要求与车辆信誉值分配自身的计算资源以执行不同的卸载任务。最后,通过实验仿真,验证了方案的有效性和可靠性。算资源以执行不同的卸载任务。最后,通过实验仿真,验证了方案的有效性和可靠性。     

A mechanism for network resource allocation and task offloading in mobile edge computing and network engineering

At present, most of the resource allocation methods in mobile edge computing allocate computing resources according to the time order in which task requests are calculated and unloaded, without considering the priority of tasks in practical applications. According to the computing requirements in such cases, a priority task-oriented resource allocation method is proposed. According to the average processing time of the task execution, the corresponding priority for task is given. The tasks with different priorities are weighted to allocate computing resources, which not only ensures that the high-priority tasks obtain sufficient computing resources, but also reduces the total time and energy consumption to complete the calculation of all tasks, thus improving the quality of service. The experimental results show that the proposed method can achieve better performance.     

基于深度强化学习的车辆边缘计算任务卸载方法

车辆边缘计算允许车辆将计算任务卸载到边缘服务器,从而满足车辆爆炸式增长的计算资源需求。但是如何进行卸载决策与计算资源分配仍然是亟待解决的关键问题。并且,运动车辆在连续时间内进行任务卸载很少被提及,尤其对车辆任务到达随机性考虑不足。针对上述问题,建立动态车辆边缘计算模型,描述为7状态2动作空间的Markov决策过程,并建立一个分布式深度强化学习模型来解决问题。另外,针对离散—连续混合决策问题导致的效果欠佳,将输入层与一阶决策网络嵌套,提出一种分阶决策的深度强化学习算法。仿真结果表明,所提算法相较于对比算法,在能耗上保持了较低水平,并且在任务完成率、时延和奖励方面都具备明显优势,这为车辆边缘计算中的卸载决策与计算资源分配问题提供了一种有效的解决方案。     

考虑任务依赖的卫星边缘计算资源分配与卸载决策算法

针对高低轨卫星网络协同边缘计算的卸载决策问题,提出了一种考虑任务依赖的联合计算资源、无线资源分配与任务调度的卫星网络边缘计算卸载决策算法。首先,将任务卸载问题建模为最小化任务延迟和能量消耗的联合优化问题;然后,将能源消耗和时延引入子任务优先级定义中,基于动态优先级进行启发式卸载策略搜索。该算法保证了子任务之间的依赖性并同时考虑了无线资源分配。仿真结果表明,与已有研究相比,该算法能缩短高低轨卫星协同计算的任务执行延迟,且能够降低低轨卫星功耗。     

MEC联合资源分配与计算卸载的效益

针对资源受限的移动边缘计算(MEC)卸载问题,提出一种基于遗传算法优化的卸载决策与计算资源分配方法(GAO).建立联合时延、能耗以及卸载费用的系统卸载效益模型,提出最小资源分配阈值;引入改进的遗传算法求解效益最大化问题,针对该问题提出一种两段式的染色体结构和遗传算子.进行仿真实验,对比分析随机卸载决策与平均计算资源分配...     

A self-learning fuzzy approach for proactive resource provisioning in cloud environment

The development of a communication infrastructure has made possible the expansion of the popular massively multiplayer online games. In these games, players all over the world can interact with one another in a virtual environment. The arrival rate of new players to the game environment causes fluctuations and players always expect services to be available and offer an acceptable service-level agreement (SLA), especially in terms of response time and cost. Cloud computing emerged in the recent years as a scalable alternative to respond to the dynamic changes of the workload. In massively multiplayer online games applications, players are allowed to lease resources from a cloud provider in an on-demand basis model. Proactive management of cloud resources in the face of workload fluctuations and dynamism upon the arrival of players are challenging issues. This paper presents a self-learning fuzzy approach for proactive resource provisioning in cloud environment, where key is to predict parameters of the probability distribution of the incoming players in each period. In addition, we propose a self-learning fuzzy autoscaling decision-maker algorithm to compute the proper number of resources to be allocated to each tier in the massively multiplayer online games by applying the predicted workload and user SLA. We evaluate the effectiveness of the proposed approach under real and synthetic workloads. The experimental results indicate that the proposed approach is able to allocate resources more efficiently than other approaches.     

移动边缘计算中资源受限的串行任务卸载策略

边缘计算通过在靠近用户的网络边缘侧部署计算和存储资源,使用户可将高延迟、高耗能应用程序卸载到网络边缘侧执行,从而降低应用延迟和本地能耗. 已有的卸载研究通常假设卸载的任务之间相互独立,且边缘服务器缓存有执行任务所需的所有服务. 然而,在真实场景中,任务之间往往存在依赖关系,且边缘服务器因其有限的存储资源只能缓存有限的服务. 为此,提出一种在边缘服务器计算资源和服务缓存有限的约束下,权衡时延和能耗（即成本）的依赖性任务卸载方法. 首先,松弛研究问题中的约束将其转换为凸优化问题;采用凸优化工具求最优解,并用解计算卸载任务的优先级. 然后,按照优先级将任务卸载到成本最小的边缘服务器,若多个依赖任务卸载到不同的边缘服务器,为了使总成本最小,则采用改进粒子群算法求解边缘服务器的最佳传输功率. 最后,为了验证所提方法的有效性,基于真实数据集进行了充分的实验. 实验结果表明,所提方法与其他方法相比能够降低总成本8%~23%.

     

移动边缘计算中基于深度强化学习的计算卸载调度方法

针对移动边缘计算中具有依赖关系的任务的卸载决策问题,提出一种基于深度强化学习的任务卸载调度方法,以最小化应用程序的执行时间。任务调度的过程被描述为一个马尔可夫决策过程,其调度策略由所提出的序列到序列深度神经网络表示,并通过近端策略优化(proximal policy optimization)方法进行训练。仿真实验表明,所提出的算法具有良好的收敛能力,并且在不同环境下的表现均优于所对比的六个基线算法,证明了该方法的有效性和可靠性。     

A review on the computation offloading approaches in mobile edge computing: A game-theoretic perspective

In recent years, novel mobile applications such as augmented reality, virtual reality, and three-dimensional gaming, running on handy mobile devices have been pervasively popular. With rapid developments of such mobile applications, decentralized mobile edge computing (MEC) as an emerging distributed computing paradigm is developed for serving them near the smart devices, usually in one hop, to meet their computation, and delay requirements. In the literature, offloading mechanisms are designed to execute such mobile applications in the MEC environments through transferring resource-intensive tasks to the MEC servers. On the other hand, due to the resource limitations, resource heterogeneity, dynamic nature, and unpredictable behavior of MEC environments, it is necessary to consider the computation offloading issues as the challenging problem in the MEC environment. However, to the best of our knowledge, despite its importance, there is not any systematic, comprehensive, and detailed survey in game theory (GT)-based computation offloading mechanisms in the MEC environment. In this article, we provide a systematic literature review on the GT-based computation offloading approaches in the MEC environment in the form of a classical taxonomy to recognize the state-of-the-art mechanisms on this important topic and to provide open issues as well. The proposed taxonomy is classified into four main fields: classical game mechanisms, auction theory, evolutionary game mechanisms, and hybrid-base game mechanisms. Next, these classes are compared with each other according to the important factors such as performance metrics, case studies, utilized techniques, and evaluation tools, and their advantages and disadvantages are discussed, as well. Finally, open issues and future uncovered or weakly covered research challenges are discussed and the survey is concluded.     

基于边缘计算的新型任务卸载与资源分配策略

针对移动边缘计算（MEC）中密集型任务卸载时，系统开销较大和延时抖动明显的问题，提出一种新型资源分配策略。首先在系统时延约束下，分析了系统任务执行开销与终端设备的资源分配机制；其次建立了基于计算卸载和任务分配的联合凸优化目标；最后采用拉格朗日乘子法进行迭代更新得到最优解。仿真结果表明，所提任务卸载与资源分配方案在保证用户服务质量的同时降低了任务执行开销，并有效提升了MEC系统性能。     

Content-centric data and computation offloading in AI-supported fog networks for next generation IoT

Fog Computing (FC) based IoT applications are encountering a bottleneck in the data management and resource optimization due to the dynamic IoT topologies, resource-limited devices, resource diversity, mismatching service quality, and complicated service offering environments. Existing problems and emerging demands of FC based IoT applications are hard to be met by traditional IP-based Internet model. Therefore, in this paper, we focus on the Content-Centric Network (CCN) model to provide more efficient, flexible, and reliable data and resource management for fog-based IoT systems. We first propose a Deep Reinforcement Learning (DRL) algorithm that jointly considers the content type and status of fog servers for content-centric data and computation offloading. Then, we introduce a novel virtual layer called FogOrch that orchestrates the management and performance requirements of fog layer resources in an efficient manner via the proposed DRL agent. To show the feasibility of FogOrch, we develop a content-centric data offloading scheme (DRLOS) based on the DRL algorithm running on FogOrch. Through extensive simulations, we evaluate the performance of DRLOS in terms of total reward, computational workload, computation cost, and delay. The results show that the proposed DRLOS is superior to existing benchmark offloading schemes.     

移动云计算中的任务调度与计算迁移算法

随着互联网的发展,许多应用程序对计算机的计算能力和资源的需求越来越大,而移动设备具有有限的资源和计算能力,云计算迁移技术是解决计算密集型任务在移动端上顺利运行的主流方法。针对无线网络中联合调度和迁移的问题,提出了一个快速高效的启发式算法。算法将能够迁移的任务全部迁移到云端作为初始解,然后逐次计算可迁移任务在移动端运行的能耗节省量,依次将节省量最大的任务迁移到移动端。每迁移一个任务,该算法都会依据任务间的通信时间,及时更新各个任务的能耗节省量。为了进一步优化启发式算法得到的解,还构造了适用于此问题并以启发解为初始解的模拟退火算法,给出了相应的编码方法、目标函数、邻域解、温度参数以及算法终止准则。与无迁移、饱和迁移、随机迁移三类算法的对比实验结果表明,由启发式算法得出的解具有高效性,能给出使移动端能耗更小的解。     

基于多智能体深度强化学习的车联网可信任务卸载策略

针对车联网中边缘节点的可信性无法保证的问题,提出了一种基于声誉的车联网可信任务卸载模型,用记录在区块链上的边缘节点声誉来评估其可信度,从而帮助终端设备选取可靠的边缘节点进行任务卸载。同时,将卸载策略建模为声誉约束下的时延和能耗最小化问题,采用多智能体深度确定性策略梯度算法来求解该NP-hard问题的近似最优解,边缘服务器依据任务卸载的完成情况获得奖励,然后据此更新记录在区块链上的声誉。仿真实验表明,与基准测试方案相比,该算法在时延和能耗方面降低了25.58%~27.44%。     

基于组合预测模型的云计算资源负载预测研究

随着云计算技术的不断发展,云计算资源负载变化呈现出越来越复杂的特征。针对云计算资源的负载预测问题,综合考虑云计算环境中资源负载时间序列的线性与非线性特性,提出了一种基于自回归移动平均模型ARIMA与长短期记忆网络LSTM的组合预测模型LACL。使用公开数据集与传统负载预测模型进行了对比实验,实验结果表明,该云计算资源组合预测模型预测精度明显高于其他预测模型,显著 降低了云环境中对资源负载的实时预测误差。     

A benefit-aware on-demand provisioning approach for multi-tier applications in cloud computing

Dynamic resource provisioning is a challenging technique to meet the service level agreement (SLA) requirements of multi-tier applications in virtualization-based cloud computing. Prior efforts have addressed this challenge based on either a cost-oblivious approach or a cost-aware approach. However, both approaches may suffer frequent SLA violations under flash crowd conditions. Because they ignore the benefit gained that a multi-tier application continuously guarantees the SLA in the new configuration. In this paper, we propose a benefit-aware approach with feedback control theory to solve this problem. Experimental results based on live workload traces show that our approach can reduce resource provisioning cost by as much as 30% compared with a costoblivious approach, and can effectively reduce SLA violations compared with a cost-aware approach.     

基于深度强化学习的移动边缘计算任务卸载研究

在移动边缘计算中,本地设备可以将任务卸载到靠近网络边缘的服务器上进行数据存储和计算处理,以此降低业务服务的延迟和功耗,因此任务卸载决策具有很大的研究价值.首先构建了大规模异构移动边缘计算中具有多服务节点和移动任务内部具有多依赖关系的卸载模型;随后结合移动边缘计算的实际应用场景,提出利用改进的深度强化学习算法优化任务卸载策略;最后通过综合比较任务卸载策略的能耗、成本、负载均衡、延迟、网络使用量和平均执行时间等指标,分析了各卸载策略的优缺点.仿真实验结果表明,基于长短期记忆(long short-term memory, LSTM)网络和事后经验回放(hindsight experience replay, HER)改进的HERDRQN算法在能耗、费用、负载均衡和延迟上都有很好的效果.另外利用各算法策略对一定数量的应用进行卸载,通过比较异构设备在不同CPU利用率下的数量分布来验证卸载策略与各评价指标之间的关系,以此证明HERDRQN算法生成的策略在解决任务卸载问题中的科学性和有效性.     

移动边缘计算环境下的动态资源分配策略

在通讯设备爆炸式增长的时代,移动边缘计算作为5G通讯技术的核心技术之一,对其进行合理的资源分配显得尤为重要。移动边缘计算的思想是把云计算中心下沉到基站部署(边缘云),使云计算中心更加靠近用户,以快速解决计算资源分配问题。但是,相对于大型的云计算中心,边缘云的计算资源有限,传统的虚拟机分配方式不足以灵活应对边缘云的计算资源分配问题。为解决此问题,提出一种根据用户综合需求变化的动态计算资源和频谱分配算法(DRFAA),采用"分治"策略,并将资源模拟成"流体"资源进行分配,以寻求较大的吞吐量和较低的传输时延。实验仿真结果显示,动态计算资源和频谱分配算法可以有效地降低用户与边缘云之间的传输时延,也可以提高边缘云的吞吐量。     

基于改进MADDPG的UAV轨迹和计算卸载联合优化算法

在地震、台风、洪水、泥石流等造成严重破坏的灾区,无人机(unmanned aerial vehicle, UAV)可以作为空中边缘服务器为地面移动终端提供服务,由于单无人机有限的计算和存储能力,难以实时满足复杂的计算密集型任务.本文首先研究了一个多无人机辅助移动边缘计算模型,并构建了数学模型;然后建立部分可观察马尔可夫决策过程,提出了基于复合优先经验回放采样方法的MADDPG算法(composite priority multi-agent deep deterministic policy gradient, CoP-MADDPG)对无人机的时延能耗以及飞行轨迹进行联合优化;最后,仿真实验结果表明,本文所提出算法的总奖励收敛速度和收敛值均优于其他基准算法,且可为90%左右的地面移动终端提供服务,证明了本文算法的有效性与实用性.