基于平均等待时间约束的松弛预留机制

传统预留机制存在预留请求拒绝率高、可能会延迟本地作业执行的缺陷。为此,提出一种基于平均等待时间(MWT)约束的松弛预留机制,并给出相应的预留接纳算法。采用松弛预留策略,在每个资源上设置任务等待时间阈值,若调度队列中排队任务的MWT超过该阈值,调度系统将使部分排队任务尽快获得空闲资源,从而减少MWT。实验结果表明,该预留机制能有效保障排队任务的MWT、提高资源利用率、降低预留请求拒绝率。     

一种基于松弛时间的服务网格资源能力预留机制

通过资源能力预留为网格服务提供确定的QoS保证是实现服务网格QoS管理的基础和关键.针对确定性资源能力预留的"资源能力碎片"问题,提出了一种支持松弛时间的灵活资源能力预留机制,并设计了支持松弛时间的资源预留请求接纳控制算法.在CROWN的节点服务器上实现了该机制,并通过仿真实验进行性能评价.结果表明,和已有的确定型预留机制相比,支持松弛时间的资源能力预留机制,使资源能力调度具有更多的自主性,可显著地提高网格资源的综合利用效率.     

基于资源-预留图的动态网格资源预留机制

在传统的静态资源预留机制(static resource reservation mechanism,简称SRRM)下,用户的预留请求一旦通过接纳测试就立即被调度到某个资源上.因此,SRRM无法考虑在预留提前时间(book-ahead time)内作业队列的变化对调度目标产生的影响以及资源故障对预留任务的影响.提出了一种动态的网格资源预留机制(dynamic resource reservation mechanism,简称DRRM),在该机制下,通过接纳测试的预留请求在其实际使用资源时才被调度.引入了资源-预留图(resource-reservation graph,简称RRG)对DRRM进行说明,并给出了RRG的修改规则.DRRM能够感知预留提前时间内资源状态的变化,根据其运行时信息动态地调度已接纳的预留请求.模拟实验结果表明,与SRRM相比,DRRM损失了一定的预留请求接纳率,但大大减少了预留任务对非预留任务的抢占,显著提高了网格资源的有效利用率,并且对网格资源故障具有更好的容错效果.     

一种可选择最佳带宽的多路路由算法

实现服务质量保证,在多用户、多请求接入的网络中都是采取为相应请求预留资源的方式,而多请求预留资源使得网络局部出现资源"碎块".单路路由算法不能利用这些资源接入用户请求.为提高网络的资源利用率,采用多路路由方法,在单一通路不能满足连接请求要求时,寻找多条能联合满足要求的一组虚拟不相交并路,并预留资源共同满足用户的服务质量要求.为验证算法的有效性,在ns2模拟器上实现了所提出的算法,并与其他算法进行了性能比较.实验结果表明,提出的算法能很好地适应网络运行状况.在网络轻负载情况下,通信开销小.在网络重负载情况下,使用并行多路增加连接请求的接通率.将此算法应用于支持服务质量保证的多媒体数据网,能充分利用有限的网络资源,提高传输多媒体信息的能力.     

跨数据中心的动态资源联合预留研究

如何有效保障服务质量(Quality of Service,QoS)是网格计算和云计算等网络计算平台面临的核心问题之一,资源可获得性和性能的动态变化使得将多媒体网络的资源提前预留机制直接应用到网络计算平台面临巨大挑战.为此提出了一种跨数据中心的资源联合预留体系架构,引入虚拟资源容器以适应资源的动态性,提出了资源预留协商算法以增加预留请求与资源的匹配率.在东南大学校园云计算平台的实验结果表明,在资源动态变化的网络计算环境中,所提出的体系架构能显著提高预留请求的成功率,且具有更强的稳定性.     

一种面向效益均衡的网格资源预留机制

网格环境中的资源预留机制为跨站点的协同资源分配提供可靠的服务质量保证.针对传统预留机制存在本地任务QoS难以得到保障的缺陷,提出一种基于效益均衡的预留机制.该机制通过比较资源顸留收益与预留对本地任务执行所造成的负面影响来确定资源预留数量,以期取得网格任务QoS保障和本地任务QoS保障的效益均衡.模拟实验采用真实系统负载在较大规模的计算网格系统中检验了该机制的性能表现,实验结果显示,基于效益均衡的预留机制能有效提高资源利用率,同时能显著降低任务的资源费用情况.     

云计算环境下动态资源碎片管理机制

针对云计算环境下用户所需资源与服务资源规格不完全相符以及在资源预留过程中完整资源被切割而产生的资源碎片问题,提出一种云环境下考虑碎片资源重利用的动态资源管理策略。研究了云计算环境下资源碎片的形成原因,构建了碎片资源池,制定了任务对碎片资源接收的度量标准,在充分考虑当前任务对资源查找、调度、匹配的同时,进一步讨论了任务调度对资源的分割情况,进而指出资源碎片对后续任务接收能力的影响,搭建了云计算环境下动态资源碎片调度模型。理论分析和Cloudsim仿真实验证明,该资源管理策略能有效实现碎片资源的优化重组,提高了资源对后续任务的接收能力,与此同时保证了较高的资源利用率。     

一种面向随机性负载的计算网格资源预留定价策略

针对传统预留机制对本地任务服务质量造成的负面影响,提出了基于预留收益与损失均衡的资源预留机制。首先,基于本地任务相关统计特性利用概率论方法,给出并证明了一段时间内本地任务执行时间之和的概率分布函数。然后,在资源提供者的效用中考虑本地任务可能带来的损失,求解出了为保障本地任务服务质量的最低资源价格。实验结果表明,提出的预留机制能有效保障本地任务的服务质量,提高资源利用率、降低任务拒绝率。     

一种可分片预留接纳控制算法研究

接纳控制算法的好坏直接影响分布式计算中资源提前预留机制的总体性能.针对现有灵活资源预留接纳控制算法的优缺点,提出了一种可分片预留接纳控制算法.当无法实现固定资源预留时,该算法在保证最大分片间隔的前提下,允许对资源进行分片预留;在各分片中,若存在剩余资源量小于请求预留资源量的时隙,允许用最小资源量进行预留.通过与3种可拓展预留接纳控制算法(缩短持续时间,增大预留带宽(shorten the duration and increase the reserved bandwidth,SDIB);减小预留带宽,延长持续时间(reduce the reserved bandwidth and extend the duration,RBED);改变预留的开始时间(change the reserved start time,CST))的对比实验,从接纳率和有效资源利用率方面进行了评估.实验结果表明,可分片预留接纳控制算法能有效减少资源碎片,具有更优的综合性能.     

基于松弛预留策略的连续双向拍卖模型

针对网格资源的松弛预留问题,提出一种改进的连续双向拍卖模型。在网格资源定价策略中,引入松弛时间保证更高的任务预留接纳率。在用户出价和要价策略中,买方通过剩余时间和剩余资源量出价,卖方根据负载情况要价。仿真实验结果表明,对于具有费用约束的网格任务,该模型能增加约21%的资源总收益,提高约15%的资源利用率。     

基于效益函数与优先级的网格服务质量保证

将资源预约请求作为网格调度对象是提高网格整体服务质量的有效手段。针对传统预约方式成功率低的问题,提出将效益函数与优先级引入资源预约,采用可连续预约的方法,在减少预约双方交互负担的同时,有效提高资源预约成功率。通过模拟与比较,表明该算法在应对大量预约请求时均有较好的性能。     

Distributed meta-scheduling in lambda grids by means of Ant Colony Optimization

Data transport infrastructures based on wavelength-routed optical networks are appealing to the grid community because they can provide a reconfigurable and high-bandwidth network service to emerging data-intensive grid applications. Traditionally, the control plane of those optical networks is independent from the management of other resources, such as computing systems, and it is commonly limited to fulfill immediate reservation demands of lightpaths. However, the lack of network integration to grid resources and the absence of advance reservation of bandwidth are critical aspects to the widespread adoption of optical networks in grid environments. Therefore, in this work, a distributed grid meta-scheduler based on an Ant Colony Optimization (ACO) algorithm is proposed, which is capable of co-allocating both optical networking and grid resources, under an integrated, extended and distributed control plane that supports advance reservations. The blocking probability and the delay for starting the processing of a request are evaluated for the proposed meta-scheduler under different resource co-allocation algorithms and different meta- and local scheduling policies, including the influence of the information aggregation across grid nodes. In addition, the benefits of the proposed approach are shown, such as grid and networking resource integration at the control plane, and capital expenditure reductions at the deployed optical network when compared to an immediate reservation scenario.     

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.     

基于间隔型双边拍卖的网格资源预留算法

针对已有基于竞价拍卖的网格资源预留方法存在的问题，提出了一种基于间隔型双边拍卖的网格资源预留算法。根据网格资源预留的需求和拍卖机制的特点，给出了网格系统模型；基于间隔型双边拍卖的思想，设计了适合网格系统的资源拍卖机制，给出了详细的网格资源预留算法；算法的最优性分析表明：在资源拍卖过程，提供真实信息对用户方和资源方均是唯一占优策略，且算法能够保证中标资源供需平衡。 最后，模拟实验结果证明了算法的有效性和正确性。     

Approximate Dynamic Programming for Stochastic Resource Allocation Problems

A stochastic resource allocation model, based on the principles of Markov decision processes (MDPs), is proposed in this paper. In particular, a general-purpose framework is developed, which takes into account resource requests for both instant and future needs. The considered framework can handle two types of reservations (i.e., specified and unspecified time interval reservation requests), and implement an overbooking business strategy to further increase business revenues. The resulting dynamic pricing problems can be regarded as sequential decision-making problems under uncertainty, which is solved by means of stochastic dynamic programming (DP) based algorithms. In this regard, Bellman’s backward principle of optimality is exploited in order to provide all the implementation mechanisms for the proposed reservation pricing algorithm. The curse of dimensionality, as the inevitable issue of the DP both for instant resource requests and future resource reservations, occurs. In particular, an approximate dynamic programming (ADP) technique based on linear function approximations is applied to solve such scalability issues. Several examples are provided to show the effectiveness of the proposed approach.      

An RSVP based seamless resource reservation scheme for wireless mobile networks

In order to achieve continuous QoS provision to a mobile user, it is required to deploy a mechanism which reserves resources in advance at the locations to which the mobile user is expected to move. An important issue with regards to these mechanisms is to minimize the waste of resources caused by aforementioned reservations while keeping up the QoS guaranteed services to the user. In this paper, we propose a wireless network domain structure which deploys a special cell, called a gray cell. The boundaries of wireless network regions are surrounded by the gray cells that belong to all of its neighboring regions. Based on this wireless domain structure, we propose a mechanism which completely removes the waste of redundant resource reservation in the wireless network domain while greatly reducing the risk of reservation disruption caused by inter-region handovers. Using simulation, it is shown that the proposed mechanism can deal with inter-region handover as effectively as it does with intra-region handover. It is also presented that the proposed mechanism outperforms existing mechanisms with respect to reservation disruption time and packet losses caused by handovers for various mobility patterns of a mobile user.     

Designing end-to-end resource reservations in predictable distributed embedded systems

Contemporary distributed embedded systems in many domains have become highly complex due to ever-increasing demand on advanced computer controlled functionality. The resource reservation techniques can be effective in lowering the software complexity, ensuring predictability and allowing flexibility during the development and execution of these systems. This paper proposes a novel end-to-end resource reservation model for distributed embedded systems. In order to support the development of predictable systems using the proposed model, the paper provides a method to design resource reservations and an end-to-end timing analysis. The reservation design can be subjected to different optimization criteria with respect to runtime footprint, overhead or performance. The paper also presents and evaluates a case study to show the usability of the proposed model, reservation design method and end-to-end timing analysis.