数据网格中服务质量感知的副本放置方法

可靠的服务质量是数据网格应用的关键因素之一.服务质量感知的副本放置问题在传统模型中加入个体服务质量约束,可满足QoS严格的数据网格应用需求.针对现有副本放置算法不支持多属性约束、可扩展性不高等缺点,提出一种基于矩阵操作加速的3阶段副本放置算法TP-GABMAC算法,并引入副本环处理副本更新和一致性维护问题.分析和实验表明,TP-GABMAC算法具有稳定性好和可扩展性强的特点,在多种网络拓扑、访问模式和负载条件下均能获得合理的副本策略.     

QoS感知的副本放置问题研究综述

副本技术是分布式存储中的重要技术.与传统副本放置问题不同,QoS感知的副本放置关注单个访问请求的服务质量需求满足.介绍了QoS感知的副本放置问题出现的背景,归纳了系统的抽象模型与问题的定义,并逐一介绍了当前几种较好的算法.通过自行设计和实现的副本放置模拟器,对现有算法进行了定性和定量分析,讨论了算法的优劣性,并指出进一步研究需要关注的问题.     

基于遗传算法的副本管理策略

在数据网格环境中,为了提高数据的可靠性和降低用户对数据访问的延迟,广泛采用了数据复制技术;由于副本技术的引入,就会面临一个很直接的问题,即一个应用程序如何根据副本性能和访问特性,从一组副本中选择一个最佳副本,就是副本选择问题.针对数据复制技术中的关键技术:副本创建和副本选择,在价格机制模型和并行数据传输的基础上,采用了遗传算法实现副本的优化管理.最后使用网格模拟器OptorSim对算法进行测试分析,结果显示基于遗传算法的策略有更好的性能.     

云存储中动态副本放置机制研究

数据副本管理是云计算系统管理的重要组成部分,在云计算系统的海量数据处理过程中,针对目前已知的数据存放与资源调度算法存在考虑副本动态性和可靠性的不足,提出了一种动态的副本放置机制。该机制基于区域结构,考虑数据处理时其副本的数量和放置位置,以及副本的产生对于内存和带宽等系统资源的开销:首先根据云存储中的副本信息,对被访问频率高且访问平均响应时间长的数据信息进行复制,并给出副本数量的计算方法;考虑缩小副本分布的节点选择范围,提出动态的副本放置算法——DRA,将一定范围内的节点根据提出的域的划分,进行放置筛选,以存放数据副本。实验结果表明,提出的动态放置机制不仅减少了低访问率副本对系统存储空间的浪费;同时也减少了高访问率副本所需跨节点的传输延迟,有效提高了云存储系统中的数据文件的访问效率、负载的均衡水平,以及云存储系统的可靠性和可用性。     

对等网络中一种优化的副本分布方法

数据复制技术是一种提高P2P系统中数据可靠性和可用性的常用策略.现有复制方法大多只考虑副本数量,副本数量越多就越能提高资源访问效率,但采用这样的数据复制方法将会带来高昂的副本一致性维护代价.为平衡副本一致性维护的开销和多副本带来的访问性能提升之间的关系,该文提出了动态副本分布方法.文中首先给出了副本目录的设计和副本信息的获取方法,能够获得某一逻辑资源的所有副本信息.然后,根据逻辑资源的全局副本信息,对访问频率高且平均响应时间长的数据资源进行复制,并给出副本数量的计算方法.最后,根据用户访问特征和节点实时带宽等信息计算放置副本的最佳地点,使副本分布能够适应数据访问请求和网络带宽的动态变化.模拟实验结果显示,该方法能够实现全局优化的副本分布,以少量数据副本提升资源访问的性能.     

数据网格中一种基于滑动窗口的副本创建和替换策略

数据网格中,数据副本技术提高了数据的访问速度,减少了带宽的消耗.副本创建策略是数据副本研究中的重要问题之一.提出了在P2P网络环境下,一种基于滑动窗口技术的副本创建和替换策略.使用传输时间比作为副本创建和替换的依据.分析和模拟显示,该方法能在控制访问空间的同时,获得良好的性能.     

数据网格环境下副本淘汰策略的研究

为了在数据网格环境中不增加副本存储空间的条件下,能够很好地进行数据副本的淘汰,提出了一种改进副本淘汰算法.该算法利用权重函数兼顾访问时间和访问频率,在考虑副本传输代价的因素上引入动态调整因子μ,根据实际情况动态的调整副本传榆代价所占的比例.仿真实验结果表明,该算法在副本尺寸差异较大的情况下,可以大大减少副本淘汰误差,提高了网格结点的作业平均执行时间和网络有效利用率.     

副本放置中的更新策略及算法

副本技术广泛应用于云计算及分布式系统中,合理的数据副本放置是降低网络运行成本的重要手段,也是副本技术的核心问题。副本更新是针对网络中数据访问请求的动态变化而进行的副本添加与删除。针对副本放置问题,提出了一种基于多访问策略的副本动态更新算法MPFSF(min_placement far servers first)。该算法在引入通信距离限制的同时,尽可能多地重用网络中已存在的副本,并根据需要实施必要的副本更新,最大使用每个副本的处理能力,以便降低数据访问的时延,提高数据访问效率。最后通过实验结果和分析证明,该算法在不同的通信距离限制下,网络的运行成本得到了明显的降低,对原有算法的改进最高可达84.6%。     

云存储智能多数据副本放置机制

数据副本管理机制是云存储系统的重要组成部分。为了提高云存储系统的可伸缩性、可靠性,同时改善用户访问时间,通常采用多数据副本机制,并且需要解决数据副本放置问题。为此,提出了一种用于云存储系统的智能多数据副本放置机制。该机制基于p-中心模型,以最小化访问代价为优化目标,基于遗传算法（genetic algorithm,GA）确定优化的数据副本放置方案,基于生物地理学优化（biogeography-based optimization,BBO）算法确定用户访问请求对数据副本的优化分配。基于CloudSim进行了仿真实现和性能评价,结果表明,云存储智能多数据副本放置机制是可行和有效的。     

基于萤火虫优化的副本放置方法

在云计算环境下分布式存储系统中,通常采用副本技术保证存储系统的可用性和可靠性,放置策略是副本技术的一个关键问题。针对现有副本放置策略中存在的副本访问开销大的问题,提出一种基于离散型萤火虫优化的副本放置算法。考虑副本放置对用户访问性能的影响,对其建立数学模型,计算萤火虫位置的适应度函数,并朝着荧光素值最大即最优值移动,进而得到合适的副本放置节点。通过仿真实验评估提出的方法,并与基于蚁群算法的副本放置策略进行比较。实验结果证明该算法能够选择合适的副本放置节点,具有较好的收敛性,并有效地降低存储系统的副本访问开销。     

一种网格服务的质量描述方法和服务匹配策略

面向服务的网格系统需要有充分的服务质量(QoS)支持，比较QoS参数进行网格服务发现与选择，尽快达成服务的匹配，是保证用户QoS需求的有效方法。为描述网格服务的质量需求，该文定义了网格服务的基本QoS参数，提出了网格服务质量数SQN的新概念，用于度量网格服务的整体量化特性。在此基础上，提出了相应的服务匹配策略和价格确定模型，加快网格服务匹配的过程，保障用户服务质量，并作了较深入的分析与讨论。     

A resource selection scheme for QoS satisfaction and load balancing in ad hoc grid

Ad hoc grids are highly heterogeneous and dynamic, in which the availability of resources and tasks may change at any time. The paper proposes a utility based resource selection scheme for QoS satisfaction and load balancing in ad hoc grid environments. The proposed scheme intends to maximize the QoS satisfaction of ad hoc grid users and support load balancing of grid resources. For each candidate ad hoc grid resource, the scheme obtains values from the computations of utility function for QoS satisfaction and benefit maximization game for ad hoc grid resource preference. The utility function for QoS satisfaction computes the utility value based on the satisfaction of QoS requirements of the grid user request. The benefit maximization game for grid resource node preference computes the preference value from the resource point of view. Its main goal is to achieve load balancing and decrease the number of resource selection failure. The utility value and the preference value of each candidate ad hoc grid resource are combined to select the most suitable grid resource for ad hoc grid user request. In the simulation, the performance evaluation of proposed algorithm for ad hoc grid is conducted.     

Joint optimisation of application QoS and energy conservation in grid environment

Mobile and wireless devices with limited energy are now of interest for the grid community. Reducing the related energy consumption is very important for wireless devices. Compared to conventional grid systems, energy aware grids need to support application quality of service (QoS) with limited energy. There is an inherent conflict in the design goals for high grid application QoS and low energy consumption. To address this challenging problem, this article considers joint optimisation of application QoS and energy conservation in grid environment. Joint optimisation of application QoS and energy conservation is targeted to maximise the system utility without exceeding the deadline and the total energy available, which can be provided by an exhaustible source such as a battery. This article formulates joint optimisation of application QoS and energy conservation as constrained maximisation problem; the constraints include energy budget and application QoS requirements. This article also presents a QoS and energy aware scheduling algorithm (QESA) which balances application QoS and energy conservation to improve system performance. In the simulation, the performance evaluation of QESA is conducted.     

Dynamic multi-resource advance reservation in grid environment

How to guarantee user’s QoS (Quality of Service) demands become increasingly important in a service-oriented grid environment. Current research on grid resource advance reservation, a well known and effective mechanism to guarantee QoS, fails to adapt to dynamic behavior of grid, and cannot solve imprecise deny of reservation request problem efficiently. For this, enabling system architecture for advance reservation is proposed. Virtual resource container (VRC) is adopted to alleviate a negative effect resulted from dynamic behavior of grid and QoS deviation distance (QDD) based logical resource selection algorithm is put forward to decrease imprecise reject ratio of reservation. At last, this new architecture is deployed to campus grid, and two illustrative experiments are conducted. Experimental results show that system architecture for advance reservation proposed in this paper can alleviate negative influence of grid resource dynamic fluctuation and avoid imprecise reject of advance reservation request effectively.     

服务网格环境下一种基于整体QoS的服务匹配策略

研究服务网格中的服务质量(QoS)保障问题.定义基本QoS参数及量化要求,阐述QoS参数归一计算方法,提出一种基于整体QoS的服务匹配策略.该策略利用QoS约束的一致性和整体QoS特性进行网格服务发现与选择,从而减小满足最小服务要求的解空间.实验结果表明该策略有效提高了匹配服务的执行效率,满足了用户的QoS需求.     

基于QoS的网格错误处理服务选择模型

在检测到错误发生的情况下,错误管理服务如何针对实际的不同网格应用程序,选择恰当的错误处理服务,是保证可靠网格容错能力的一个关键问题.针对网格计算环境下的错误处理需求,提出一种基于服务质量QoS的错误处理服务选择模型.在分析网格错误处理的相关背景及需求基础上,给出几种常用错误处理技术的形式化定义,提出相应的服务质量标准;将基于QoS的错误处理服务选择问题抽象为多属性决策问题,建立相应的决策模型.     

基于UML的网格QoS面向方面建模方法

利用基于UML的面向方面编程(AOP)技术来建模实时系统,结合传统的网格QoS结构模型,把QoS关注从系统中抽象为一个独立于系统的QoS方面,从结构建模、行为建模、方面织入以及代码产生几个方面实现面向方面的建模方法,更好地提供网格环境下QoS保证,分离了QoS关注,提高软件的重用性.     

Cross-layer optimization policy for QoS scheduling in computational grid

This paper presents a cross-layer quality of service (QoS) optimization policy for computational grid. Efficient QoS management is critical for computational grid to meet heterogeneity and dynamics of resources and users’ requirements. There are different QoS metrics at different layers of computational grid. To improve perceived QoS by end users over computational grid, QoS supports can be addressed in different layers, including application layer, collective layer, fabric layer and so forth. The paper tackles cross-layer grid QoS optimization as optimization decomposition, each layer corresponds to a decomposed subproblem. The proposed policy produces an optimal set of grid resources, service compositions and user's payments at the fabric layer, collective layer and application layer respectively to maximize global grid QoS. The cross-layer optimization problem decomposes into three subproblems: grid resource allocation problem, service composing and user satisfaction degree maximization problem, all of which interact through the optimal variables for capacities of grid resources and service demand. In order to coordinate the subproblems, cross-layer QoS feedback mechanism is established to ensure different layer interactions. The simulations are conducted to validate the efficiency of the proposed policy.     

An adaptive algorithm for QoS-aware service composition in grid environments

Service composition enables users to realize their complex needs as a single request and it has been recognized as a flexible way for resource sharing and application integration since the appearance of Service-Oriented Architecture (SOA). For each of the needed individual services there may be many candidate services available presented by different vendors and with different functional and non-functional properties such as Quality of Service (QoS). Approaches are needed to select candidate services with various QoS levels according to user’s performance requirements meanwhile adapt to dynamic churn in grid environments. This paper mainly focuses on adaptive management of QoS-aware service composition in grid environments and proposes an adaptive algorithm for QoS-aware service composition (AQSC). In AQSC we model this problem as the Multi-Constrained Optimal Path selection problem (MCOP) and use heuristic approach for service selection, then backup services set is introduced as an adaptive mechanism so as to ensure the fulfillment of composite service when some candidate services fail or withdraw. Both theoretical analysis and simulation results indicate that AQSC has high composition success rate, finish rate and low cost.