移动云计算中基于SLA的动态带宽分配方法研究与实现

发布时间：2018-08-05 15:15

【摘要】：云计算(Cloud Computing)作为一种能随时随地地访问共享资源池的全新的计算模式,已受到了广泛的关注。随着云计算和移动互联网的不断发展,支持安卓和IOS等智能设备的云计算服务已经出现,移动云计算技术应运而生并成为当今学术界一个新的研究热点。目前云计算的研究大多基于计算能力、内存、CPU利用率等参数来分配资源以提高资源利用率。然而,随着移动云计算的出现,网络带宽逐渐成为影响移动云计算技术发展的一个重要因素,对数据中心网络的资源共享起着重要作用。同时,随着移动终端用户对带宽的需求越来越高,当多个移动终端用户同时发送请求时,移动云中心服务器的性能也将大大降低。本文在对移动云计算和SLA相关技术分析和总结的基础上,针对移动云计算中不同终端用户对带宽需求的动态性以及多样性等问题,提出了一种基于SLA的动态带宽分配方法,该方法将移动终端用户请求带宽与SLA定义的协商带宽分配区间进行相似性匹配,并根据匹配结果进行等级映射,为不同等级的移动终端用户进行动态带宽分配。本文的主要工作内容如下:(1)提出了一种适合移动云计算的动态带宽分配方法。该方法采用等级映射和带宽预留的方式,将云计算中心预留的带宽分为H、M和L三个等级,然后根据各个等级的预留带宽资源是否满足移动终端用户的带宽请求情况,在移动云中心实现带宽的动态分配。(2)提出了基于SLA的动态带宽分配实现框架。在该框架中,通过WSLA语言实现相似性匹配和动态带宽分配两个模块,其中相似性匹配模块根据移动终端用户请求带宽与SLA协商签订的带宽分配区间之间的相似性匹配算法,计算出匹配结果并进行等级映射;动态带宽分配模块则根据等级映射结果为不同的移动终端用户进行动态带宽分配。(3)通过云计算仿真模拟器CloudSim,仿真实现了移动云计算中基于SLA的动态带宽分配方法原型系统。实验结果分析表明,本文提出的移动云计算中基于SLA的动态带宽分配方法在一定程度上提高了带宽资源利用率和系统吞吐量,降低了时延,同时还保证了服务质量(QoS)。
[Abstract]:Cloud computing (Cloud Computing), as a new computing mode which can access the pool of shared resources anytime and anywhere, has been paid more and more attention. With the continuous development of cloud computing and mobile Internet, cloud computing services supporting smart devices such as Android and IOS have emerged. Mobile cloud computing technology has emerged as the times require and has become a new research hotspot in academia. At present, most of cloud computing research is based on computing power, memory CPU utilization and other parameters to allocate resources to improve resource utilization. However, with the emergence of mobile cloud computing, network bandwidth gradually becomes an important factor affecting the development of mobile cloud computing technology, and plays an important role in resource sharing of data center network. At the same time, with the increasing demand for bandwidth by mobile end users, the performance of mobile cloud central servers will be greatly reduced when multiple mobile terminal users simultaneously send requests. Based on the analysis and summary of the related technologies of mobile cloud computing and SLA, this paper proposes a dynamic bandwidth allocation method based on SLA, aiming at the dynamic and diversity of bandwidth requirements of different end users in mobile cloud computing. The method matches the request bandwidth of mobile terminal user with the negotiated bandwidth allocation interval defined by SLA, and maps the mobile terminal bandwidth according to the matching result to make dynamic bandwidth allocation for mobile terminal users of different levels. The main work of this paper is as follows: (1) A dynamic bandwidth allocation method for mobile cloud computing is proposed. The method uses hierarchical mapping and bandwidth reservation to divide the bandwidth reserved by cloud computing center into three levels: Hum M and L, and then according to whether the reserved bandwidth resources of each level satisfy the bandwidth request of the mobile terminal user. Dynamic bandwidth allocation in mobile cloud center is realized. (2) A dynamic bandwidth allocation framework based on SLA is proposed. In this framework, two modules, similarity matching and dynamic bandwidth allocation, are implemented by WSLA language. The similarity matching algorithm is based on the similarity matching algorithm between the bandwidth request of mobile terminal user and the bandwidth allocation interval negotiated by SLA. The matching result is calculated and the rank mapping is carried out. The dynamic bandwidth allocation module performs dynamic bandwidth allocation for different mobile end users according to the hierarchical mapping results. (3) the dynamic bandwidth allocation prototype system based on SLA in mobile cloud computing is implemented by cloud computing simulation simulator CloudSim. The experimental results show that the proposed dynamic bandwidth allocation method based on SLA in mobile cloud computing can improve bandwidth resource utilization and system throughput to a certain extent, reduce delay and ensure quality of service (QoS).
【学位授予单位】：江苏大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP393.09

【相似文献】