Spark Streaming动态资源分配策略

针对Spark Streaming作为混合大数据计算平台流处理组件时资源调整周期长和不能满足多应用多用户个性化需求的问题,提出了一种多应用下动态资源分配策略（DRAM）。该策略增加了应用全局变量来控制动态资源分配过程。首先,获取历史执行数据反馈和应用全局变量;然后,进行资源增减计算;最后,进行资源增减执行。实验结果表明,所提策略能够有效调整应用资源配额,且在稳定数据流和不稳定数据流两种情况下,其处理延时相比原Spark平台的Streaming策略和Core策略都有所降低;同时该策略也能够提高集群资源利用率。     

云计算环境下的服务调度和资源调度研究

云计算中的服务调度与资源调度对云计算的性能有重要影响,在分析现有云计算调度模式的基础上,针对云计算数据密集与计算密集的特点,提出分层调度策略以实现云计算中的服务与资源调度。分层调度策略对任务进行划分确定作业优先级,并通过数据局部性和总任务完成率对资源进行分配。数值评价部分应用分层调度与已有调度进行比较。实验结果表明,所采用的调度有效提高了资源利用率,为云服务的进一步研究提供了思路。     

IaaS环境下多租户安全资源分配算法和安全服务调度框架

针对基础设施即服务（IaaS）环境下多租户使用安全服务时由于安全资源有限和安全资源分配不均导致的效率低下问题,提出了一个租户安全资源调度框架。首先以最小最大公平算法为基础,结合Fair Scheduler的调度思想为租户设定了最小共享量和资源需求量属性;然后通过安全服务资源分配算法在保证租户最小共享量满足的前提下,尽可能公平地满足租户的资源需求;最后结合租户内任务调度和租户间资源抢占算法,实现了租户安全服务调度框架。实验结果表明,在随机资源分配条件下,安全服务资源分配算法与传统资源分配算法相比在资源利用率和作业效率上均有明显提高,安全服务调度框架可以有效解决多租户安全资源的分配和强占问题。     

面向运行时间预测和容错感知的网格资源分配

提出与描述了一种面向任务运行时间预测和容错感知（Fault-Aware）的网格资源分配策略,采用主动容错的方式,在资源出错之前尽量提前避免它出错或异常的情况发生。该策略把网格中任务的运行时间（runtime）预测和资源的在线时间（uptime）预测结合起来,相对于普通的调度策略具有比较高的资源利用率。在具体的CoBRA网格中间件中实现了该容错感知调度,描述了实现该容错感知调度策略模块的功能。测试过程中选择了睡眠任务技术,划分四种不同的场景进行实验,把该容错感知资源分配与普通的FCFS调度策略进行比较,结果证明在可变化的资源可用性的情况下系统可以加快应用的整体执行时间,具有很小的偏差。     

Resource brokering using a multi‐site resource allocation strategy for computational grids

Grid computing employs heterogeneous resources which may be installed on different platforms, hardware/software, computer architectures, and perhaps using different computer languages to solve large‐scale computational problems. As many more Grids are being developed worldwide, the number of multi‐institutional collaborations is growing rapidly. However, to realize Grid computing's full potential, it is expected that Grid participants must be able to share one another's resources. This paper presents a resource broker that employs the multi‐site resource allocation (MSRA) strategy and the dynamic domain‐based network information model that we propose to allocate Grid resources to submitted jobs, where the Grid resources may be dispersed at different sites, and owned and governed by different organizations or institutes. The jobs and resources may also belong to different clusters/sites. Resource statuses collected by the Ganglia, and network bandwidths gathered by the Network Weather Service, are both considered in the proposed scheduling approach. A dynamic domain‐based model for network information measurement is also proposed to choose the most appropriate resources that meet the jobs' execution requirements. Experimental results show that MSRA outperformed the other tested strategies. Copyright © 2010 John Wiley & Sons, Ltd.     

Semantic link‐based constructing of multi‐dimensional service resource space for collaboration

Automatic service collaboration calls for the development of semantically structured service resource space to maximize the utility of Web services. Semantic links contain rich semantic information that may indicate important relationships among services. We provided an effective method for constructing multi‐dimensional service resource space based on semantic links for service collaboration, in which similar and related semantic relationship between services are considered. We first clustered services with similar and related relations on the basis of a hierarchical structure respectively and then took advantage of the resource space model to construct multi‐dimensional service resource space. Finally, experimental results show the effectiveness of the method. Concurrency and Computation: Practice and Experience, 2012.© 2012 Wiley Periodicals, Inc.     

多业务流量预测动态带宽分配方法

静态带宽资源分配方法或基于业务分类的动态带宽分配方法等很难适应业务流特性,无法保证各级业务的服务质量。设计了流量预测及带宽分配模型,给出了流量预测计算方法,提出多业务流量预测动态带宽分配方法（MSTPDBA）,该方法基于业务预测流量的大小进行可用带宽的重新分配。仿真实验证明了MSTPDBA的有效性,能控制不同业务的时延,并提高了系统带宽资源的利用率。     

基于鲁棒优化的云医疗资源配置问题

医疗资源配置优化是云医疗系统高效运行的核心决策,然而,由于这种新型互联网医疗服务系统具有多组织协同、上下转诊以及诊疗时间不确定等特点,上述问题可以描述为需求不确定情形下核心医生服务时间分配优化问题.构建一个以最小化最大医疗服务成本为目标函数的云医疗资源鲁棒配置优化模型,通过引入决策者对患者诊疗时间和转诊概率两种不确定性...     

Buffer and server allocation in general multi‐server queueing networks

This paper deals with the joint optimization of the number of buffers and servers, an important issue since buffers and servers represent a significant amount of investment for many companies. The joint buffer and server optimization problem (BCAP) is a non‐linear optimization problem with integer decision variables. The performance of the BCAP is evaluated by a combination of a two‐moment approximation (developed for the performance analysis of finite general‐service queues) and the generalized expansion method (a well‐known method for performance analysis of acyclic networks of finite queues). A standard non‐linear optimization package is used to optimize the BCAP for a large number of experiments. A comprehensive set of numerical results is presented. The results show that the methodology is capable of handling the trade‐off between the number of servers and buffers, yielding better throughputs than previously published studies. Also, the importance of the squared coefficient of variation of the service time is stressed, since it strongly influences the approximate optimal allocation.     

Energy‐efficient task allocation with quality of service provisioning for concurrent applications in multi‐functional wireless sensor network systems

Multi‐functional wireless sensor network (WSN) system is a new design trend of WSNs, which are evolving from dedicated application‐specific systems to an integrated infrastructure that supports the execution of multiple concurrent applications. Such system offers inherent advantages in terms of cost and flexibility because it allows the effective utilization of available sensors and resource sharing among multiple applications. However, sensor nodes are very constrained in resources, mainly regarding their energy. Therefore, the usage of such resources needs to be carefully managed, and the sharing with several applications imposes new challenges in achieving energy efficiency in these networks. In order to exploit the full potential of multi‐functional WSN systems, it is crucial to design mechanisms that effectively allocate tasks onto sensors so that the entire system lifetime is maximized while meeting various application requirements. However, it is likely that the requirements of different applications cannot be simultaneously met. In this paper, we present the Multi‐Application Requirements Aware and Energy Efficiency algorithm as a new resource allocation heuristic for multi‐functional WSN system to maximize system lifetime subject to various application requirements. The heuristic effectively deals with different quality of service parameters (possibly conflicting) trading those parameters and exploiting heterogeneity of multiple WSNs. Copyright © 2013 John Wiley & Sons, Ltd.     

一种基于流量预测的资源动态管理算法

分析了基于有效带宽理论(effectivebandwidththeory)的静态资源分配方案的弊端。在此基础上,提出了一种基于流量预测的资源动态管理算法,并且把资源动态管理算法具体应用到QoS的区分服务(DifferentiatedService)体系中,算法在区分服务网络的边缘路由器上实现。最后,在ns 2的仿真环境下对两种算法进行了比较,试验结果证明无论在丢包率还是链路利用率上,资源动态管理算法都比静态资源分配方案有明显的优势。     

云计算环境下多DAG调度的资源分配进化算法

现有多DAG调度研究主要在多个DAG共享资源调度的时间最小化、公平性最大化、吞吐量最大化等问题方面提出了相关的解决方案,然而,现有的方法还不能很好地解决云计算环境下多DAG共享资源调度的资源分配优化问题.为此,首先分析讨论了一组多DAG共享云计算资源调度中的多DAG数量、属性结构分布特点与资源需求量之间的关系,并在此基础上提出了基于资源需求强度预测变异方法的进化算法EFRD,有效地解决了云计算环境下多DAG共享资源调度的资源分配优化问题,既保证了多DAG的调度执行时间最小化,也避免了资源的浪费.实验表明,EFRD算法能够有效地收敛到最优解.     

移动边缘计算中资源受限的动态服务部署策略

移动边缘计算（MEC）的出现使移动用户能够以低延迟访问部署在边缘服务器上的服务。然而,MEC仍然存在各种挑战,尤其是服务部署问题。边缘服务器的数量和资源通常是有限的,只能部署数量有限的服务;此外,用户的移动性改变了不同服务在不同地区的流行度。在这种情况下,为动态请求部署合适的服务就成为一个关键问题。针对该问题,通过了解动态用户请求来部署适当的服务以最小化交互延迟,将服务部署问题表述为一个全局优化问题,并提出了一种基于集群划分的资源聚合算法,从而在计算、带宽等资源约束下初步部署合适的服务。此外,考虑动态用户请求对服务流行度及边缘服务器负载的影响,开发了动态调整算法来更新现有服务,以确保服务质量（QoS）始终满足用户期望。通过一系列仿真实验验证了所提出策略的性能。仿真结果表明,与现有基准算法相比,所提出的策略可以降低服务交互延迟并实现更稳定的负载均衡。     

无线异构网络的资源分配策略

无线异构网络要求实现不同工作模式通信系统间的资源合理分配。为了获得不同通信系统对无线资源的有效利用,参考XG架构定义了可承载不同通信系统业务需求的多维度资源容器,提出二级资源分配策略及相应的资源分配算法来实现多维度资源容器与各个通信系统之间各种业务承载需求的匹配。仿真与性能分析验证了该资源分配策略在实现不同通信系统共存的条件下,提高了无线网络频谱资源利用率。     

一种改进的Parlay多业务环境资源分配算法

针对Parlay多业务环境下由软交换业务发起的业务请求和由Web服务器发起的业务请求,提出了一种改进的自适应资源分配算法。该算法通过识别不同业务负荷的轻重程度,自动调整为其分配的资源数目,在保证一定的公平性和系统稳定的前提下,使资源得到充分的利用。仿真结果表明,这种自适应的资源分配算法能有效地满足下一代网络多业务环境中各业务的需要。     

基于软件体系结构和广义差分进化的测试资源动态分配算法

测试资源分配是软件测试中的一个基础问题,然而已有研究大都针对并串联模块软件模型而鲜有涉及体系结构软件模型。为此,首先针对可靠性和错误数动态变化的测试环境,构建了一种基于体系结构的多阶段多目标测试资源分配模型。然后基于参数重估计、种群重新初始化、广义差分进化和归一化加权求和设计了一种面向动态可靠性和错误数的多阶段多目标测试资源分配算法。最后,在仿真实验中,与已有的归一化加权求和多目标差分进化（WNS-MODE）算法进行对比,所提算法在不同结构的体系结构软件模型实例上所获解集更优,容量值提高了约16倍,覆盖值提高了约84个百分点,超体积提高了约6倍。实验结果表明,所提算法能够更好地适应可靠性和错误数的动态变化,可为体系结构软件模型的动态测试提供更多和更优的测试资源分配方案。     

A self‐scalable load injection service

Load testing of applications is an important and costly activity for software provider companies. Classical solutions are very difficult to set up statically, and their cost is prohibitive in terms of both human and hardware resources. Virtualized cloud computing platforms provide new opportunities for stressing an application's scalability, by providing a large range of flexible and less expensive (pay‐per‐use model) computation units. On the basis of these advantages, load testing solutions could be provided on demand in the cloud. This paper describes a Benchmark‐as‐a‐Service solution that automatically scales the load injection platform and facilitates its setup according to load profiles. Our approach is based on: (i) virtualization of the benchmarking platform to create self‐scaling injectors; (ii) online calibration to characterize the injector's capacity and impact on the benched application; and (iii) a provisioning solution to appropriately scale the load injection platform ahead of time. We also report experiments on a benchmark illustrating the benefits of this system in terms of cost and resource reductions. Copyright © 2013 John Wiley & Sons, Ltd.     

A novel model on dynamic resource allocation in optical networks

A novel model on dynamic resource allocation in the WDM optical networks is proposed, basing on the integrated considerations of the impacts of transmission impairments and service classification on dynamic resource allocation in the optical layer. In this model, the priorities of optical connection requests are mapped into different thresholds of transmission impairments, and a uniform method which is adopted to evaluate the virtual wavelength path (VWP) candidates is defined. The Advanced Preferred Wavelength Sets Algorithm (A-PWS) and the heuristic Dynamic Min-Cost & Optical Virtual Wavelength Path Algorithm (DMC-OVWP) are presented addressing the routing and wavelength assignment (RWA) problem based on dynamic traffic and multi priorities in wavelength-routed optical networks. For a received optical connection request, DMC-OVWP is employed to calculate a list of the VWP candidates, and an appropriate VWP which matches the request's priority is picked up to establish the lightpath by analyzing the tra     

Distributed optimal resource allocation of second‐order multiagent systems

In this paper, the resource allocation problems of multiagent systems are investigated. Different from the well‐studied resource allocation problems, the dynamics of agents are taken into account in our problem, which results that the problem could not be solved by most of existing resource allocation algorithms. Here, the agents are in the form of second‐order dynamics, which causes the difficulties in designing and analyzing distributed resource allocation algorithms. Based on gradient descent and state feedback, two distributed resource allocation algorithms are proposed to achieve the optimal allocation, and their convergence are analyzed by constructing suitable Lyapunov functions. One of the two algorithms can ensure that the decisions of all agents asymptotically converge to the exact optimal solution, and the other algorithm achieves the exponential convergence. Finally, numerical examples about the economic dispatch problems of power grids are given to verify the effectiveness of the obtained results.