排队系统定价控制问题的研究综述

定价作为服务系统价值因素的调控机制,在拥塞控制、收益管理和服务质量控制等领域有重要的应用.服务提供商设计定价策略使得系统性能在诱导的顾客行为下达到最优.排队模型是分析服务系统性能的重要模型.本文介绍了排队系统定价控制的基本要素,包括排队模型、顾客行为、性能指标以及定价策略类型,围绕静态定价策略、动态定价策略和优先级定价策略,详细综述了排队系统定价控制的研究现状,然后简要介绍了作者针对排队网络动态定价问题的最新研究进展,最后总结了该领域的研究现状、尚未解决的问题、当前的研究热点以及未来的发展方向.     

多实例云计算资源市场下超额预订决策方法

针对现有云供应商数据中心负载率低、云用户需求不确定及多样性的问题,为提高云供应商平均利润,建立了不确定需求下的多实例类型云服务超额预订模型。该模型结合实际云计算资源市场下超额预订对于云供应商负载均衡及云服务等级协议(SLA)的影响,给出超额预订的多重约束条件,提出了各实例类型数量最优分配策略。实验结果表明,采用该模型在预约未使用概率为0.25时,云供应商利润较高,数据中心负载率达到78%,最终确定了各实例类型的最优分配数量。     

云图书馆中“基础设施即服务”模式的安全研究

云计算环境下的数字图书馆具有支持大量资源共享、具备强大计算能力等特点,但同时也因其支持多用户共享资源而带来了诸如存储数据被还原篡改、信息隐私泄露、通信数据被窃听等等安全问题。其中,以虚拟化技术为关键技术的"基础设施即服务"(IaaS)模式亦如此。鉴于此,论文从服务提供商、租赁用户、服务提供商和租赁用户间供求关系的角度出发,探讨了云图书馆服务架构中IaaS模式的资源共享安全问题,并提出了相应的安全策略。     

多云环境下基于博弈论的用户行为分析模型

针对多云环境下用户的分布式拒绝服务攻击缺乏有效处理机制的现状,从云服务提供商收益角度出发,提出多云环境下基于博弈论的用户行为分析模型。模型首先基于博弈论构造收益矩阵,之后利用模糊隶属度函数判定用户的行为,并进一步评估非协作和协作场景下云服务提供商的资源消耗和收益。经仿真验证,协作模型能够在减少资源消耗的基础上,有效地降低云服务提供商遭受分布式拒绝服务攻击的风险,相对于非协作场景,可以将单位资源的收益提高3倍以上,具有很强的现实意义。     

面向云联网的云服务协商机制

随着云计算的快速发展,越来越多的用户开始使用云服务提供商提供的服务,而云联网作为云计算研究的新领域,可以实现跨云服务提供商的服务,当单个云服务提供商无法满足用户的服务需求时,云服务提供商之间以合作的方式为用户提供服务,以便更好地满足用户的服务需求。针对上述情况,提出了面向云联网的云服务协商机制,该机制利用云联网和改进的经典合同网模型来实现云服务提供商的交互协商。为了有效地选出合作伙伴以提高合作效率,还为每一个云服务提供商建立了一个熟人集。实验表明,本文设计的机制可以有效地提高云服务提供商之间的合作效率,并且可以更好地满足用户的服务需求。     

基于队列理论的云资源分配收益最大化算法

为了优化资源分配收益,提出基于时间服务因子TSF资源定价机制下的资源分配算法。首先,算法将资源分配问题形式化为队列模型,并构建了收益最大化函数。然后,通过Lagrange乘子法求解最优化函数的解。利用时间服务因子,算法分别通过安全满意度因子ASF和响应满意度因子RSF定义了资源的定价方式,并同步考虑定价、请求到达率、资源服务率及可用资源量,得到了使收益最大化的资源分配方式。实验结果表明,与传统的启发式资源分配算法比较,该算法得到的收益更高,且尤其在云资源数量较稀少的场景下,算法将更加具有优势。     

基于服务级别协议的云资源分配

与网格、集群等传统计算模式不同,云计算为用户提供了一种利用远程计算资源的实用商业模型。在不同的客户之间动态分配云资源池以获得最大收入,成为云服务提供商最为关心的问题。云计算中心需要把面向客户的服务层指标转换为面向系统的操作层指标,根据服务级别协议动态管理云计算资源。研究了基于服务级别协议的服务提供商收入最大化问题,借助排队论模型对资源分配问题进行了形式化描述,然后依据定价机制、服务请求到达率、服务率、可用资源等因素给出了资源最优分配方案。实验结果表明,该算法优于相关算法。     

CDNs中基于反馈控制的资源分配控制器设计

当前的内容分发网络配置策略既无法应付不可预知的突发访问事件,又会导致网络利用率较低,浪费大量资源;云计算可以针对用户需求配置资源,并且为服务采取即收即付定价策略;基于云技术的自适应视频流分发网络提出一种控制机制,同时采取反馈控制技术,设计了一种动态资源分配控制器,通过调节云技术CDN网络中的虚拟机数量,在向用户提供最高质量的视频服务的同时,实现传输成本最小化;实验结果表明,相对静态控制器、前馈控制器和"有RAC无SP"控制器,控制器的分发成本分别下降33%、28%、10%。     

云计算IaaS现货实例定价方法研究

目前云计算IaaS具有三种定价模式-按需运行、保留定制、现货竞价,前两种模式价格相对稳定,现货竞价模式的价格却处于不断波动中,目前还没有相应的云计算IaaS定价标准,价格的高低直接影响资源利用率和提供商利润.针对该问题,提出一种云服务提供商利润最大化目标下的云计算IaaS定价方法.考虑价格因素对现货竞价市场需求和整个云计算市场需求的双重影响,利用价格需求弹性理论,以价格手段间接控制市场需求;针对现货竞价的市场风险,引入机会成本;运用数学模型求解出最优定价区间,通过算例验证了价格变动对市场需求和利润的影响,并分析了需求价格弹性对利润的敏感性.此研究对云计算提供商的定价策略具有指导意义.     

A game-theoretic approach to the financial benefits of infrastructure-as-a-service

Financial benefits are an important factor when cloud infrastructure is considered to meet processing demand. The dynamics of on-demand pricing and service usage are investigated in a two-stage game model for a monopoly Infrastructure-as-a-Service (IaaS) market. The possibility of hybrid clouds (public clouds plus own infrastructure) turns out to be essential in order that not only the provider but also the clients have significant benefits from on-demand services. Even if the client meets all demand in the public cloud, the threat of building a hybrid cloud keeps the instance price low. This is not the case when reserved instances are offered as well. Parameters like load profiles and economies of scale have a huge effect on likely future pricing and on a cost-optimal split-up of client demand between either a client’s own data center and a public cloud service or between reserved and on-demand cloud instances.     

Two pricing mechanisms for a service provider when customers’ delay costs are value-related

In this paper, we study two pricing mechanisms for a provider that serves delay-sensitive customers, one is the uniform pricing and the other is the priority auction. The expected delay cost of a customer is assumed to depend on his value for the service (i.e. the unit delay cost is a strictly increasing function of his value) and the expected sojourn time caused by the number of customers in the system (and his payment-based position in the queue if auction is adopted). Hence, each customer reacts to the service provider’s pricing mechanism by deciding whether or not to enter the service system and how much he pays. This resulting problem is a Stackelberg game. When auction is adopted, by using of adverse selection, we derive a feasible scheme in which customers with higher value would like to pay more. We further compare the performance of these two pricing mechanisms. Our numerical examples show that auction performs better not only in terms of revenue making but also in terms of social welfare improvement. Interestingly, auction can also render more customer surplus in most instances, which differs from the common techniques in revenue management field.     

云计算中TSP问题求解服务的定价机制

旅行商问题(TSP)是一个典型的路径优化问题,在城市交通规划、物流运输、通信网络设置等领域都存在类似的问题和应用。但是,TSP问题的求解是NP难的,当问题规模很大时,必须借助大规模并行计算环境,例如云计算平台,以较大的计算开销来获得可行解。以TSP问题为具体实例,研究云计算服务的定价机制。一般情况下,定价机制要满足公平、灵活、动态、自适应。从公平合理角度来看,影响计算服务定价的因素主要有两方面:一是求解问题的难度,包括计算时间复杂性、空间复杂性、输入输出数据规模等;二是求解服务质量,即服务契约,包括可以作为服务等级协定指标的求解精度、响应时间、资源要求等。由此,提出了一种新的云计算中的服务定价机制:CloudPricing。该机制给出了服务定价的一般和具体原则,并给出了相应的定价公式。针对TSP问题求解,进行了具体的定价实例分析,这对云计算中NP难问题求解服务的定价有参考意义。     

Framework for cloud-based software test data generation service

This paper presents the framework of cloud-based software test data generation service (CSTS) that caters to cost-effective test data generation service in a cloud environment. In contrast to existing conventional or cloud-based testing frameworks, CSTS has a number of unique benefits. First, CSTS is designed to facilitate test data generation in minimum time and cost. Second, unlike existing frameworks which mandates clients to opt for resources to test their jobs, CSTS guides customer for selecting best cluster configuration in order to minimize the cost. While the existing models do not provide any solution for trust establishment in cloud computing services, CSTS delivers it by implementing security mechanism with the provision of role based access control. The security mechanism proposed in this paper ensures the protection of data and code of different users. Third, CSTS provides a mathematical pricing model to fulfill the expectations of customers and also to maximize the net profit of service providers. Cloud service request model has also been designed that postulates service level agreements between customers and service providers. We have evaluated, compared, and analyzed our framework and have found that it outperforms other existing cloud-based frameworks.     

Resource Management in Cloud Computing with Optimal Pricing Policies

As a new computing paradigm, cloud computing has received much attention from research and economics fields in recent years. Cloud resources can be priced according to several pricing options in cloud markets. Usage-based and reserved pricing schemes are commonly adopted by leading cloud service providers (CSPs) such as Amazon and Google. With more and more CSPs entering cloud computing markets, the pricing of cloud resources is an important issue that they need to consider. In this paper, we study how to segment cloud resources using hybrid pricing schemes in order to obtain the maximum revenue by means of optimal pricing schemes in what is a largely monopolized cloud market. We first study how the revenue of a cloud provider can be maximised using an on-demand pricing scheme. We then turn to the study of revenue maximization with a reserved pricing scheme and, finally, we compare the revenues obtained from the two pricing schemes.     

Private cloud for collaboration and e-Learning services: from IaaS to SaaS

The idea behind cloud computing is to deliver Infrastructure-, Platform- and Software-as-a-Service (IaaS, PaaS and SaaS) over the Internet on an easy pay-per-use business model. The Hochschule Furtwangen University (HFU) is running their own private cloud infrastructure, called Cloud Infrastructure and Application CloudIA. The targeted users of the CloudIA project are HFU staff and students running e-Learning applications, and external people for collaboration purposes. Therefore, in this paper, we introduce our work in building a private cloud. More specifically, this paper shows how our cloud offerings in each of the cloud service models, i.e. IaaS, PaaS and SaaS, address the requirements and needs of e-Learning and collaboration in an university environment.     

Enhancing Federated Cloud Management with an Integrated Service Monitoring Approach

Cloud Computing enables the construction and the provisioning of virtualized service-based applications in a simple and cost effective outsourcing to dynamic service environments. Cloud Federations envisage a distributed, heterogeneous environment consisting of various cloud infrastructures by aggregating different IaaS provider capabilities coming from both the commercial and the academic area. In this paper, we introduce a federated cloud management solution that operates the federation through utilizing cloud-brokers for various IaaS providers. In order to enable an enhanced provider selection and inter-cloud service executions, an integrated monitoring approach is proposed which is capable of measuring the availability and reliability of the provisioned services in different providers. To this end, a minimal metric monitoring service has been designed and used together with a service monitoring solution to measure cloud performance. The transparent and cost effective operation on commercial clouds and the capability to simultaneously monitor both private and public clouds were the major design goals of this integrated cloud monitoring approach. Finally, the evaluation of our proposed solution is presented on different private IaaS systems participating in federations.