A game-theoretic method of fair resource allocation for cloud computing services

As cloud-based services become more numerous and dynamic, resource provisioning becomes more and more challenging. A QoS constrained resource allocation problem is considered in this paper, in which service demanders intend to solve sophisticated parallel computing problem by requesting the usage of resources across a cloud-based network, and a cost of each computational service depends on the amount of computation. Game theory is used to solve the problem of resource allocation. A practical approximated solution with the following two steps is proposed. First, each participant solves its optimal problem independently, without consideration of the multiplexing of resource assignments. A Binary Integer Programming method is proposed to solve the independent optimization. Second, an evolutionary mechanism is designed, which changes multiplexed strategies of the initial optimal solutions of different participants with minimizing their efficiency losses. The algorithms in the evolutionary mechanism take both optimization and fairness into account. It is demonstrated that Nash equilibrium always exists if the resource allocation game has feasible solutions.     

Grid resource management based on economic mechanisms

Resources in the grid context belong to different control organizations with different interest, therefore the economic interest of each grid participant should be considered. The economic mechanism can guarantee the interest of participants in the grid with fairness and efficiency. In this paper, an economic-based resource management framework is put forward for grid computing, and then how to determine the price of resources with the economic mechanism is studied. A general equilibrium method is presented for general resources and a double auction method is proposed for special resources in the grid environment, respectively. Simulations are performed and experimental results indicate that the two methods are effective for corresponding application scenarios. Expanded version of a paper presented at CCGrid 2005.     

基于拍卖机制的网格资源分配方法

网格资源的动态性、异构性、自治性等特点,使得网格资源分配成为一个难点。目前存在的大多数分配方法仅关注分配效率,却对提高资源分配的公平性缺乏深入的研究。针对此问题,提出一种基于拍卖机制的网格资源分配方法,利用资源分配比例的算法分配资源。通过仿真实验表明该方法适合网格系统中的资源分配,能有效分配资源,提高了资源利用率,同时资源分配的公平性也得到显著的提高。     

基于效益博弈的云计算资源动态可协调分配策略研究

在对用户的任务进行计算资源分配时,为了有效提高计算资源的利用效率,减少任务执行所需要的成本,提出了一种基于效益博弈的云计算资源动态可协调分配机制。该机制采用时间矩阵和费用矩阵作为任务效益的衡量指标,提出效益博弈模型,通过该模型的效益计算方程来得到最好的资源分配策略。为了使得计算资源能够合理地按需进行分配,提出了动态可协调分配机制,在合理地分配资源,满足所有任务正常执行时所需资源的同时,最大化任务的执行效益。实验仿真及对比结果表明,在任务完成时间、任务执行的平均成本、任务完成成功率上,本文算法都取得了较好的效果。     

Workload management of cooperatively federated computing clusters

Cooperative resource sharing enables distinct organizations to form a federation of computing resources. The motivation behind cooperation is that organizations are likely to serve each other by trading unused CPU cycles given the existence of irregular usage patterns of their local resources. In this way, resource sharing would enable organizations to purchase resources at a feasible level while meeting peak computational throughput requirements. This federation results in community grid that must be managed. A functional broker is deployed to facilitate remote resource access within the community grid. A major issue is the problem of correlations in job arrivals caused by seasonal usage and/or coincident resource usage demand patterns. These correlations incur high levels of burstiness in job arrivals causing the job queue of the broker to grow to an extent such that its performance becomes severely impaired. Since job arrivals cannot be controlled, management strategies must be employed to admit jobs in a manner that can sustain a fair level of resource allocation performance at all participating organizations in the community. In this paper, we present a theoretical analysis of the problem of job traffic burstiness on resource allocation performance in order to elicit the general job management strategies to be employed. Based on the analysis, we define and justify a job management strategies for the resource broker to cope with overload conditions caused by job arrival correlations.     

Ex-post efficient resource allocation for Self-organizing Cloud

Self-organizing Cloud is a scalable model to provide powerful computability with distributed computers. The resource allocation on it is very challenging since it not only involves various types of divisible resources but needs to cope with social competitions. We propose a novel resource allocation scheme with three features on ex-post efficiency. (1) Ex-post win–win effect: each participant (consumers and suppliers) should always feel satisfied with their ex-post payoffs. (2) Incentive compatibility: we can prove each rational participant gets its optimal payoff iff their resource demands and expected prices are truthfully declared. (3) Ex-post maximized efficiency: more powerful resources should be consumed with higher likelihood, such that the whole system runs quite efficiently with maximized resource utilization. Our simulation shows the approach significantly improves resource suppliers’ incomes by 20% compared to their expectations. Meanwhile, we can guarantee consumers’ payments below their budgets, with no any degradation of task execution performance.     

适应异构集群的Mesos多资源调度DRF增强算法

云计算集群环境下多资源分配的公平性是考量资源调度子系统最重要的指标之一,DRF作为通用的多资源公平分配算法,在异构异质的集群环境下可能有失公平性。在研究Mesos框架中DRF多资源公平分配算法的基础上,设计并实现了增加机器性能评估影响因子的meDRF分配算法。将计算节点的机器性能得分,作为DRF主导份额计算的因子,使得计算任务有均等的机会获得优质计算资源和劣质计算资源。通过选取K-means、Bayes及PageRank等多种作业进行实验,实验结果表明:meDRF较DRF分配算法更能体现多资源分配的公平性,且资源分配具有更好的稳定性,能有效提高系统资源的利用率。     

Customer oriented resource allocation framework in cognitive radio

Spectrum is among the most heavily regulated and expensive resource around the world. In an effort to improve the efficiency of its usage, alternative spectrum allocation models are being proposed. One of these approaches is the Dynamic Spectrum Access which allows the cognitive radio to use or share the spectrum in an opportunistic manner. This paper proposes a framework for an architecture where unutilized bands are relicensed to different service providers on a short term basis. This framework focuses on the resource allocation process of these leased bands to customers. The proposed model could be utilized as a decision support tool which both satisfies the quality of service requirements of individual users and achieve an efficient utilization of network resources. The hypothesis is that customer information yields an efficient customer-oriented resource management strategy; therefore the allocation process integrates the information about the customer demographics and usage behavior. Simulations with real-life scenarios in cellular networks have been performed: handoff decision, resource allocation at call initiation and profit maximization in allocation of mobile base stations. In each case, the results show that the proposed model performs better in terms of revenue increase when compared to the First-Come First-Serve based approach. The simulation results prove that the population size of a customer group is not always an indicator for higher revenue for a service provider.     

基于QoE的LTE多业务资源分配算法

高效地利用无线频谱资源和保证用户体验质量是未来无线网络的主要目标。基于此，提出一种基于QoE的LTE多业务资源分配算法。在考虑信道信息、QoS要求及公平性的基础上，引入QoE来计算的用户优先级。特别的，引入最小QoE约束来保证RT用户QoE要求；提出一种次优资源块（Resource Block，RB）分配算法来解决复杂的资源分配优化问题，该算法主要分为两步：保证RT用户最小QoE要求；最大化系统加权和速率。仿真结果表明，相较现有的RT/NRT资源分配算法，该算法在用户分组丢失率、平均QoE和小区频谱效率方面性能都有所提升。     

Analysis of the Fairness of Grid Resource Allocation Based on Multicommodity Market Model

Fairness is one kind of the requirements in QoS of grid services. It is also important to optimize the allocation of grid resources and keep the overall stability of grid systems. At present, more and more economic models are applied to the grid resource management. In this paper, the fairness of grid resource allocation based on multicommodity market model is studied. Some definitions and criteria of the fairness are presented, and the fairness mechanisms of grid resource allocation are also proposed.     

负载均衡的主导资源公平分配算法

随着使用云计算并行且可靠地处理计算问题成为一种趋势,各种云计算平台应运而生,在这些平台中,保证多种资源调度策略的公平性非常重要。主导资源公平分配算法DRF有效地实现了多种资源环境中的公平分配,但在资源分配过程中容易出现集群负载不均的情况。因此,提出在使用DRF算法分配资源过程中,通过集群中各节点的资源利用率情况对节点进行K-means聚类分析,根据聚类结果将资源分配给任务来提高集群负载均衡的能力。基于CloudSim 4.0实现了改进DRF算法的仿真实验,实验结果表明,负载均衡的DRF算法比原始的DRF算法以及基于层次分析法(AHP)改进的DRF算法更能有效地改善集群整体的负载均衡。     

Fair and distributed peer-to-peer allocation of a common, refillable resource

We consider the general problem of distributed and fair peer-to-peer (P2P) allocation of a common, refillable resource. This problem recurs in a number of scenarios, for example grid computing, content distribution, Internet Service Provider service sharing, and distributed file storage over asymmetric channels. We present several distributed schemes for this allocation problem and show that these schemes guarantee two key properties: (i) asymptotic fairness, in that (even maliciously colluding) users are proportionally assigned resources corresponding to what they contribute; (ii) natural incentive to join and cooperate fairly in the system. We demonstrate the practicability of our approaches on a prototype P2P file storage system designed for typical residential Internet connections, in which download capacities often significantly exceed upload capacities. Our implementation shares file data when communications are idle using random linear codes so that, when needed, an end-user can download a file from several sources at a higher data rate than his home computer’s upload capacity. We present experimental results that support our analytical guarantees.     

Dynamic QoS queuing control mechanism for multimedia differentiated services

An essential goal of communication networks is to provide multimedia services with QoS streaming. A properly designed multimedia QoS system must reserve requested resources according to user QoS requirements and the available network resources. However, the static resource allocation among priority queues in DiffServ networks leads to insufficient resource usage when a burst occurs in one priority queue while other queues starve. This study presents a User-Oriented QoS Streaming System to achieve perceptible satisfaction based on novel streaming and media differentiation policies in DiffServ networks. This study also proposes that the Dynamic QoS Queue Mapping (DQ²M) mechanism dynamically control queue scheduling by adaptively maximizing the utilization of queues and network resources according to the soft states of the DiffServ network. Evaluation results indicate that the proposed DQ²M algorithm can improve the fairness and efficiency of resource utilization for low-priority queues.     

CA-DAG: Modeling Communication-Aware Applications for Scheduling in Cloud Computing

This paper addresses performance issues of resource allocation in cloud computing. We review requirements of different cloud applications and identify the need of considering communication processes explicitly and equally to the computing tasks. Following this observation, we propose a new communication-aware model of cloud computing applications, called CA-DAG. This model is based on Directed Acyclic Graphs that in addition to computing vertices include separate vertices to represent communications. Such a representation allows making separate resource allocation decisions: assigning processors to handle computing jobs, and network resources for information transmissions. The proposed CA-DAG model creates space for optimization of a number of existing solutions to resource allocation and for developing novel scheduling schemes of improved efficiency.     

Resource virtualization methodology for on-demand allocation in cloud computing systems

The resources’ heterogeneity and unbalanced capability, together with the diversity of resource requirements in cloud computing systems, have produced great contradictions between resources’ tight coupling characteristics and user’s multi-granularities requirements. We propose a resource virtualization model and its on-demand allocation oriented infrastructure mainly providing computing services to solve that problem. A loosely coupled resource environment centered on resource users is created to complete a mapping from physical view of resources to logic view of resources. Heuristic resource combination algorithm (HRCA) is proposed to transform physical resources to logic resources, which meets two requirements: randomness in combination and fluctuation control to the size of resources granularities. On the basis of the appraisal indexes presented for the on-demand allocation, resource matching algorithm (RMA), targeting at resource satisfaction with the highest resource utilization, is designed to reuse resources. RMA can satisfy users’ requirement in limited time and keep resource satisfaction in the highest level in the condition of logic resources granularities being less than their required size. Resource reconfiguration algorithm (RRA) is presented to implement resource matching in the condition that virtual computing resource pool cannot match granularities of resource requirements. RRA assures the lowest resource refusal rate and the greatest resource satisfaction. We verify the effectiveness, performance and accuracy of algorithms in implementing the goal of resource virtualization centered on resource users and on-demand allocation.     

一种云环境中的动态细粒度资源调度方法

云计算平台中普遍采用固定资源量的粗粒度资源分配方式,由此会引起资源碎片、过度分配、低集群资源利用率等问题.针对此问题,提出一种细粒度资源调度方法,该方法根据相似任务运行时信息推测任务资源需求;将任务划分为若干执行阶段,分阶段匹配资源,从分配时间和分配资源量两方面细化资源分配粒度;资源匹配过程中,基于资源可压缩特性进一步提高资源利用率和性能;采用资源监控、策略调整、约束检查等机制保证资源使用效率和负载性能.在开源云资源管理平台中,基于细粒度资源调度方法实现了调度器.实验结果表明：细粒度资源调度方法可以在不丧失公平性且调度响应时间可接受的前提下,细化资源匹配的粒度,有效提高云计算平台资源利用率和性能.     

Digital twin-driven service model and optimal allocation of manufacturing resources in shared manufacturing

The sharing economy has been recognized a mutually beneficial economic mode. Deriving from the concept of sharing economy, shared manufacturing was proposed under the support of advanced information and manufacturing technologies. As a core part of implementing shared manufacturing, manufacturing resource allocation aims to coordinate cross-organizational resources to provide on-demand services for personalized manufacturing requirements. However, some challenges still hinder effective and efficient resource allocation in shared manufacturing. Traditional centralized optimization methods with only one decision model are difficult to maintain autonomous decision rights of resource providers. Thus, they could hardly adapt to the situation of cross-organizational resource coordination. In addition, the credit of resource providers is rarely considered in the resource allocation process, which is unfavorable for promoting more reliable trades in shared manufacturing. To address these issues, this study proposes an integrated architecture to promote the resource allocation in shared manufacturing. A digital twin-driven service model is built to perform the seamless monitoring and control of shared manufacturing resources. The resource allocation model is constructed based on the consideration of the credit of resource providers. To keep the decision autonomy of resource providers, augment Lagrangian coordination is adopted to analyze the constructed resource allocation model. A case study is further employed to validate the effectiveness and efficiency of the proposed method in performing the resource allocation in shared manufacturing.     

异构云系统中基于智能优化算法的多维资源公平分配

资源分配策略的研究一直是云计算领域研究的热点和难点,针对异构云计算环境下多维资源的公平分配问题,结合基因算法（GA）和差分进化算法（DE）,分别给出了两种兼顾分配公平性和效率的资源分配策略,改进了解矩阵表达式使异构云系统中的主资源公平分配（DRFH）模型转化成为整数线性规划（ILP）模型,并提出了基于最大任务数匹配值（MTM）的初始解产生机制和使不可行解转化为可行解的修正操作,以此提高算法的收敛速度,使其能够快速有效地得到最优分配方案。实验结果表明,基于GA和DE算法的多维资源公平分配策略可以得到近似最优解,在最大化最小主资源份额目标值和资源利用率方面明显优于Best-Fit DRFH和Distributed-DRFH,而且针对不同任务类型的资源需求,具有较强的自适应能力。     

云环境下基于偏好的资源公平分配策略

资源分配策略是云计算领域的一个重要研究热点,其主要目标是同时考虑云用户和云提供商双方的利益,有效满足系统用户和任务的公平性,同时尽可能达到系统资源的充分利用。考虑到云环境中的用户需求各异,每个用户的任务请求数量不同,各个任务的资源需求也不同,设计了一种基于偏好的公平分配策略FABP,并给出了用户优先级和任务优先级的定义。实验分析表明,该算法不仅能缩短平均任务调度时间,而且还可以保证任务调度过程中用户和任务的公平性,实现综合资源利用率的最大化。     

Job scheduling algorithm based on Berger model in cloud environment

Considered the commercialization and the virtualization characteristics of cloud computing, the paper proposed for the first time an algorithm of job scheduling based on Berger model. In the job scheduling process, the algorithm establishes dual fairness constraint. The first constraint is to classify user tasks by QoS preferences, and establish the general expectation function in accordance with the classification of tasks to restrain the fairness of the resources in selection process. The second constraint is to define resource fairness justice function to judge the fairness of the resources allocation. We have expanded simulation platform CloudSim, and have implemented the job scheduling algorithm proposed in this paper. The experimental results show that the algorithm can effectively execute the user tasks and manifests better fairness.