基于服务等级协议的云服务成本计算模型研究
发布时间：2019-05-20 17:43
【摘要】：云计算是一种新的商业模式,而云服务则最能体现云计算的商业价值。在云服务过程中,云服务所需要支付的费用从本质上决定了服务提供商的市场竞争能力。因此,云服务提供商要保持其服务的信誉度和市场占有率,从根本上来说,则需要提高服务质量。而云服务质量最主要的衡量标准则是用户的满意度和服务条款的实施力度。 服务等级协议(Service Level Agreement, SLA)是服务提供商与用户之间签订的一份合同,合同的内容主要包括服务的内容、服务双方的职责和义务以及服务定价等相关内容。用户与提供商签订SLA,就能获得SLA中规定的服务质量,或在没有获得规定的服务质量时,便能获得提供商给予的赔偿,使服务质量有保障。云服务等级协议(Cloud Computing Service Level Agreement, SLA)是云服务提供商与用户签订的合同,用于说明服务在可用性、性能和其他可测量目标方面期望达到的服务等级,确保在合同期限内,双方约定的服务质量水平能正确执行和实现。因此,在服务质量方面,云服务提供商所提供的服务等级及其服务质量的承诺,是确保客户所使用的云服务的质量是否达到所支付水平的一项衡量标准。 目前主要的云服务提供商提供的云服务都需要按照不同的服务等级支付相应的费用,因此不同的用户将按照不同的服务内容和等级支付其相应的费用。提供商一般会综合考虑IT资源、每个部门费用支出、盈利能力和费用分配的情况,建立一个计量服务并对服务进行计费的价格模型,将服务出租给客户使用,并在SLA中明确收费标准,避免相互间就收费而产生误解和纠纷。 本文正是基于这样的现状,研究在以云SLA为引导的前提下,如何更好地规划云用户对虚拟资源的请求规则,同时减少SLA违规事例的发生；如何基于云SLA更好地规划资源分配策略来满足用户动态变化的需求,以实现虚拟资源成本最小化、SLA违规事例最小化为目标的云服务定价策略。 本文综合采用了文献调研法、模型转换方法、计算机建模方法和实证研究的方法,其内容包括六部分,第一部分(即引言)。第二部分(即第一章)简要介绍服务质量理论和SLA管理理论。第三部分(即第二章)详细研究了云SLA的相关内容。第四部分(即第三章)根据典型云服务提供商的SLA参数通过算法匹配实验得出了云SLA的关键参数指标。第五部分从云提供商的角度(第四章和第五章)和图书馆云服务的角度(第六章)分别讨论了基于SLA的云服务成本计算模型。第六部分(第七章)是研究总结和展望。各部分具体内容及研究过程如下： 第一部分首先分析了课题研究的背景及其研究意义；然后分析总结云计算、SLA以及云服务成本研究现状,通过研究现状发现：目前研究缺乏基于SLA的动态资源分配方案和请求规则来满足用户动态变化需求的成本计算模型等方面的研究；最后根据待研究的问题制定了研究思路和研究的相关内容。 第二部分介绍了服务质量QoS的相关理论和SLA的相关内容和其管理内容。 第三部分主要研究了云SLA的相关内容,构建了基于eTOM的云SLA管理模型和业务关系模型。采用这一管理模型为指导,可以为云SLA关键参数的确定提供参考依据,同时为服务过程中违规情况提供较好的解决方案。 第四部分采用本体描述语言和算法匹配的方法确定了云SLA的关键参数指标,为后文中成本计算模型的构建提供参数依据。通过实验得出云SLA的关键参数分别是：响应时间,虚拟器类型,数据传输速度,处理速度,存储能力和服务可用性。 第五部分构建了云服务的成本计算模型。首先,按照云提供商在提供云服务过程中角色的不同,本文的研究视角从云资源提供商(IaaS)和云服务提供商(SaaS/PaaS)两个角度加以开展。在IaaS服务中,研究思路是：创新性地以SLA为管理方式,研究如何合理安排用户请求的资源分配方式；研究目的是：以最小化云虚拟资源成本和SLA违规成本为原则,探讨IaaS云服务的资源请求分配规则和成本数学模型。在SaaS和PaaS服务中,基于SLA研究如何更好地规划资源分配策略来满足用户动态变化需求的数学模型和算法模型,最后通过仿真实验证明模型的合理性。其次,在继承IaaS、SaaS/PaaS成本计算模型研究成果的基础上,针对图书馆云服务的特性,研究了基于SLA的图书馆云服务成本计算模型。 本文的研究脉络是以云QoS为出发点,提出SLA可以用来保证其QoS,进而详细研究云SLA的相关内容。然后以云SLA的六个关键参数指标为基准,从虚拟资源请求规则、用户需求动态变化、虚拟资源成本、违规成本等角度,研究了基于SLA的IaaS、SaaS/PaaS和图书馆云服务的成本计算模型。
[Abstract]:Cloud computing is a new business model, and cloud services can best reflect the business value of cloud computing. In the process of cloud service, the cost of the cloud service is essentially determined by the service provider's market competitiveness. As a result, cloud service providers need to maintain the reputation and market share of their services and, fundamentally, need to improve the quality of service. The most important measure of cloud service quality is the user's satisfaction and the implementation of the service terms. The Service Level Agreement (SLA) is a contract signed between the service provider and the user. The contents of the contract mainly include the contents of the service, the responsibilities and obligations of the service parties, and the service pricing. On the other hand, when the user enters into an SLA with the provider, the service quality specified in the SLA can be obtained, or when the specified service quality is not obtained, the compensation provided by the provider can be obtained, so that the service quality is guaranteed. The Cloud Computing Service Level Agreement (SLA) is a contract between the cloud service provider and the user to illustrate the level of service that the service expects to reach in terms of availability, performance, and other measurable objectives to ensure the duration of the contract The quality of service agreed by both parties can be carried out correctly and in real time. Now, in terms of quality of service, the level of service provided by the cloud service provider and the commitment to the quality of service are a measure to ensure that the quality of the cloud services used by the customer meets the level of payment Quasi. Currently, the cloud services provided by the main cloud service providers need to pay the corresponding fees according to different service levels, so different users will pay their corresponding services according to different service content and levels The cost of the provider will generally take into account the IT resource, cost of each department, profitability and cost allocation, establish a measurement service and charge the service, lease the service to the customer for use, and explicitly charge in the SLA The standard of fee is to avoid the misunderstanding of the charges. This paper is based on the present situation, on the premise of the guidance of the cloud SLA, how to better plan the request rule of the cloud user to the virtual resource, and reduce the SLA violation. The occurrence of the case; how to better plan the resource allocation strategy based on the cloud SLA to meet the needs of the dynamic change of the user, so as to realize the virtual resource cost minimization, and the SLA violation case is minimized to the target cloud service The paper comprehensively uses the literature research method, the model conversion method, the computer modeling method and the empirical research method, the content includes six parts, the first in part (i. e., introduction), that second part (i. e., chap. I) briefly describes the quality of service theory and SLA management theory. The third part (i.e., chapter II) studies the cloud S in detail The relevant content of the LA. The fourth part (that is, Chapter III) obtains the cloud SLA through the algorithm matching experiment according to the SLA parameters of the typical cloud service provider. The fifth part discusses the cloud service based on the SLA from the perspective of the cloud provider (Chapter 4 and Chapter 5) and the angle of the library cloud service (Chapter 6). The cost-based costing model. Part VI (Chapter VII) Summary and Outlook of the study. Specific content of each section The research process is as follows: The first part first analyzes the background and the research significance of the subject research, and then analyzes the current situation of the cloud computing, the SLA and the cloud service cost. In this paper, it is found that the current research is lack of the SLA-based dynamic resource allocation scheme and the request rule to meet the requirements of the user's dynamic changes. Finally, the research is made according to the problem to be studied. The second part introduces the related theory and SLA of quality of service QoS The third part mainly studies the related content of the cloud SLA, and constructs the cloud S based on eTOM LA management model and business relationship model. With this management model as a guide, it can provide the reference basis for the determination of the key parameters of the cloud SLA, while serving the service process The fourth part uses the method of ontology description language and algorithm to determine the key parameters of the cloud SLA. The key parameters of the cloud SLA are the response time, the type of the virtual device, the data transmission speed, Processing speed, storage capacity, and service availability. The fifth part has built the costing model of the cloud service. First, according to the different roles of the cloud provider in providing the cloud service, the research view of this paper is from the cloud resource provider (IaaS) and the cloud service provider (S IaaS/ PaaS). In the IaaS service, the research approach is to study how to reasonably arrange the resource allocation method of the user's request with the SLA as the management mode; the research objective is to minimize the cloud The principles of virtual resource cost and SLA violation cost are discussed, and IaaS cloud service is discussed. The resource request allocation rule and cost mathematical model of the traffic. In the SaaS and PaaS service, the mathematical model and the algorithm model of the dynamic change demand of the user can be met based on the SLA study how to better plan the resource allocation strategy Finally, the rationality of the model is proved by the simulation experiment. Secondly, on the basis of the research results of the IaaS and the SaaS/ PaaS costing model, the paper studies the base of the library cloud service. The paper is based on the cloud QoS, and the proposed SLA can be used to ensure that the SLA can be used to guarantee the service cost. QoS, and then detail the relevant contents of the cloud SLA. Then, based on the six key parameter indexes of the cloud SLA, the IaaS, SaaS, and the SaaS based on the SLA are studied from the aspects of the virtual resource request rule, the user demand dynamic change, the virtual resource cost, the violation cost, and the like.
【学位授予单位】：武汉大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TP393.09;G350
，
本文编号：2481826
[Abstract]:Cloud computing is a new business model, and cloud services can best reflect the business value of cloud computing. In the process of cloud service, the cost of the cloud service is essentially determined by the service provider's market competitiveness. As a result, cloud service providers need to maintain the reputation and market share of their services and, fundamentally, need to improve the quality of service. The most important measure of cloud service quality is the user's satisfaction and the implementation of the service terms. The Service Level Agreement (SLA) is a contract signed between the service provider and the user. The contents of the contract mainly include the contents of the service, the responsibilities and obligations of the service parties, and the service pricing. On the other hand, when the user enters into an SLA with the provider, the service quality specified in the SLA can be obtained, or when the specified service quality is not obtained, the compensation provided by the provider can be obtained, so that the service quality is guaranteed. The Cloud Computing Service Level Agreement (SLA) is a contract between the cloud service provider and the user to illustrate the level of service that the service expects to reach in terms of availability, performance, and other measurable objectives to ensure the duration of the contract The quality of service agreed by both parties can be carried out correctly and in real time. Now, in terms of quality of service, the level of service provided by the cloud service provider and the commitment to the quality of service are a measure to ensure that the quality of the cloud services used by the customer meets the level of payment Quasi. Currently, the cloud services provided by the main cloud service providers need to pay the corresponding fees according to different service levels, so different users will pay their corresponding services according to different service content and levels The cost of the provider will generally take into account the IT resource, cost of each department, profitability and cost allocation, establish a measurement service and charge the service, lease the service to the customer for use, and explicitly charge in the SLA The standard of fee is to avoid the misunderstanding of the charges. This paper is based on the present situation, on the premise of the guidance of the cloud SLA, how to better plan the request rule of the cloud user to the virtual resource, and reduce the SLA violation. The occurrence of the case; how to better plan the resource allocation strategy based on the cloud SLA to meet the needs of the dynamic change of the user, so as to realize the virtual resource cost minimization, and the SLA violation case is minimized to the target cloud service The paper comprehensively uses the literature research method, the model conversion method, the computer modeling method and the empirical research method, the content includes six parts, the first in part (i. e., introduction), that second part (i. e., chap. I) briefly describes the quality of service theory and SLA management theory. The third part (i.e., chapter II) studies the cloud S in detail The relevant content of the LA. The fourth part (that is, Chapter III) obtains the cloud SLA through the algorithm matching experiment according to the SLA parameters of the typical cloud service provider. The fifth part discusses the cloud service based on the SLA from the perspective of the cloud provider (Chapter 4 and Chapter 5) and the angle of the library cloud service (Chapter 6). The cost-based costing model. Part VI (Chapter VII) Summary and Outlook of the study. Specific content of each section The research process is as follows: The first part first analyzes the background and the research significance of the subject research, and then analyzes the current situation of the cloud computing, the SLA and the cloud service cost. In this paper, it is found that the current research is lack of the SLA-based dynamic resource allocation scheme and the request rule to meet the requirements of the user's dynamic changes. Finally, the research is made according to the problem to be studied. The second part introduces the related theory and SLA of quality of service QoS The third part mainly studies the related content of the cloud SLA, and constructs the cloud S based on eTOM LA management model and business relationship model. With this management model as a guide, it can provide the reference basis for the determination of the key parameters of the cloud SLA, while serving the service process The fourth part uses the method of ontology description language and algorithm to determine the key parameters of the cloud SLA. The key parameters of the cloud SLA are the response time, the type of the virtual device, the data transmission speed, Processing speed, storage capacity, and service availability. The fifth part has built the costing model of the cloud service. First, according to the different roles of the cloud provider in providing the cloud service, the research view of this paper is from the cloud resource provider (IaaS) and the cloud service provider (S IaaS/ PaaS). In the IaaS service, the research approach is to study how to reasonably arrange the resource allocation method of the user's request with the SLA as the management mode; the research objective is to minimize the cloud The principles of virtual resource cost and SLA violation cost are discussed, and IaaS cloud service is discussed. The resource request allocation rule and cost mathematical model of the traffic. In the SaaS and PaaS service, the mathematical model and the algorithm model of the dynamic change demand of the user can be met based on the SLA study how to better plan the resource allocation strategy Finally, the rationality of the model is proved by the simulation experiment. Secondly, on the basis of the research results of the IaaS and the SaaS/ PaaS costing model, the paper studies the base of the library cloud service. The paper is based on the cloud QoS, and the proposed SLA can be used to ensure that the SLA can be used to guarantee the service cost. QoS, and then detail the relevant contents of the cloud SLA. Then, based on the six key parameter indexes of the cloud SLA, the IaaS, SaaS, and the SaaS based on the SLA are studied from the aspects of the virtual resource request rule, the user demand dynamic change, the virtual resource cost, the violation cost, and the like.
【学位授予单位】：武汉大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TP393.09;G350
，
本文编号：2481826
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2481826.html
