KOALA: a co‐allocating grid scheduler

In multicluster systems, and more generally in grids, jobs may require co‐allocation, that is, the simultaneous or coordinated access of single applications to resources of possibly multiple types in multiple locations managed by different resource managers. Co‐allocation presents new challenges to resource management in grids, such as locating sufficient resources in geographically distributed sites, allocating and managing resources in multiple, possibly heterogeneous sites for single applications, and coordinating the execution of single jobs at multiple sites. Moreover, as single jobs now may have to rely on multiple resource managers, co‐allocation introduces reliability problems. In this paper, we present the design and implementation of a co‐allocating grid scheduler named KOALA that meets these co‐allocation challenges. In addition, we report on the results of an analysis of the performance in our multicluster testbed of the co‐allocation policies built into KOALA . We also include the results of a performance and reliability test of KOALA while our testbed was unstable. Copyright © 2007 John Wiley & Sons, Ltd.     

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。     

A distributed multiple dimensional QoS constrained resource scheduling optimization policy in computational grid

This paper is to solve efficient QoS based resource scheduling in computational grid. It defines a set of QoS dimensions with utility function for each dimensions, uses a market model for distributed optimization to maximize the global utility. The user specifies its requirement by a utility function. A utility function can be specified for each QoS dimension. In the grid, grid task agent acted as consumer pay for the grid resource and resource providers get profits from task agents. The task agent' utility can then be defined as a weighted sum of single-dimensional QoS utility function. QoS based grid resource scheduling optimization is decomposed to two subproblems: joint optimization of resource user and resource provider in grid market. An iterative multiple QoS scheduling algorithm that is used to perform optimal multiple QoS based resource scheduling. The grid users propose payment for the resource providers, while the resource providers set a price for each resource. The experiments show that optimal QoS based resource scheduling involves less overhead and leads to more efficient resource allocation than no optimal resource allocation.     

志愿计算中基于超边际分析的资源分配方法

针对志愿计算中资源分配的特点,结合新兴古典经济学中的动态分析理论,提出了一种基于超边际分析的志愿计算资源分配方法,并使用比较优势对资源分配进行调整。模型适用于志愿计算环境中多个计算节点分别执行多种服务的情况,针对网络条件变化动态地进行响应、调整分配策略。     

On the Application of Game-Theoretic Mechanism Design for Resource Allocation in Multimedia Systems

In this paper, we study the system-level computational resource allocation problem among multiple multimedia tasks. We consider the multimedia tasks to be autonomous, i.e., they are selfish and behave strategically. We propose a resource allocation framework based on mechanism design to prevent the tasks from behaving strategically and manipulating the available system resources. We apply two mechanisms in the framework and assess their advantages over proportional-share resource allocation algorithms, which are often used in multimedia systems. We show in the simulations that the incorporation of mechanism design for system resource allocation is a promising solution that achieves efficient, fair and robust allocation against manipulation from strategic applications.      

Credit-based scheme for security-aware and fairness-aware resource allocation in cloud computing

Cloud computing systems include different types of participants with varied requirements for resources and multiple tasks; these varying requirements must be considered in the design of fairness-aware resource allocation schemes for better resources sharing. However, some participants may be malicious with a goal to damage the resource allocation fairness and increase their own utility. Hence, the resource scheduling policy must guarantee allocation fairness among the participants; further, it must ensure that fairness is not affected by the malicious usage of resources, that could cause resource exhaustion and lead to denial of service. In order to address this challenge, we propose a credit-based mechanism for resource allocation that will avoid the malicious usage of resources and, simultaneously, guarantee allocation fairness. In our scheme, a credit factor is introduced for each participant in order to evaluate the history of resource utilization and determine future resource allocation. Our model encourages a participant to release the occupied resources in timely manner after the completion of a task and imposes a punishment for malicious occupation of resources. We prove the fairness of our model and provide linear and variable gradient approaches to determine the credit factor for different scenarios. We simulate our model and perform experiments on a real cloud computing platform. The results prove the rationality, effectiveness and correctness of our approaches.     

无线MIMO系统的上行资源分配算法研究

在存在同信道干扰的无线MIMO系统中,为具有多种QoS需求的调度业务分配资源是一个具有挑战性的问题.提出一种实用的、基于SDMA的贪婪资源分配(SGRA)算法.在高效的干扰管理基础上,SGRA算法可以执行两阶段启发式计算和搜索.在第1阶段,包括上行调度和子信道分配的贪婪资源分配首先在时域频域二维进行;在第2阶段,资源分配被扩展到时域频域空域三维进行.SGRA的算法复杂度低,适用于实际无线通信系统.仿真结果表明,与同类算法相比,SGRA算法可以提高系统吞吐量,更好地保证实时业务的时延和最小数据速率需求,同时兼顾系统公平性.     

Utility-driven solution for optimal resource allocation in computational grid

Optimal resource allocation is a complex undertaking due to large-scale heterogeneity present in computational grid. Traditionally, the decision based on certain cost functions has been used in allocating grid resource as a standard method that does not take resource access cost into consideration. In this paper, the utility function is presented as a promising method for grid resource allocation. To tackle the issue of heterogeneous demand, the user's preference is represented by utility function, which is driven by a user-centric scheme rather than system-centric parameters adopted by cost functions. The goal of each grid user is to maximize its own utility under different constraints. In order to allocate a common resource to multiple bidding users, the optimal solution is achieved by searching the equilibrium point of resource price such that the total demand for a resource exactly equals the total amount available to generate a set of optimal user bids. The experiments run on a Java-based discrete-event grid simulation toolkit called GridSim are made to study characteristics of the utility-driven resource allocation strategy under different constraints. Results show that utility optimization under budget constraint outperforms deadline constraint in terms of time spent, whereas deadline constraint outperforms budget constraint in terms of cost spent. The conclusion indicates that the utility-driven method is a very potential candidate for the optimal resource allocation in computational grid.     

Resource allocation in cube network systems based on the coveringradius

When multiple copies of a certain resource exist in a cube network system, it is desirable that every nonresource node can reach the resource in a given number of hops. In this paper, we introduce systematic approaches to resource allocation in a cube system so that each nonresource node is connected with a specified number of resource copies and that the allocation performance measure of interest is optimized. The methodology used is based on the covering radius results of known codes. These codes aid in constructing desired linear codes whose codewords address nodes where resource copies are placed. The resource allocation problem is translated to an integer nonlinear program whose best possible solution can be identified quickly by taking advantage of basic properties derived from the known codes, yielding an optimal or near-optimal allocation result. Those basic properties lead to drastic time complexity reduction (up to several orders of magnitude smaller), in particular for large system sizes. Our approaches are applicable to any cube size, often arriving at more efficient allocation outcomes than what are attainable using prior schemes     

基于Agent联盟机制的网格资源协同分配

在网格计算环境，许多应用的执行经常需要同时协同分配多个资源以满足性能需求。文中提出了一种新的基于Agent联盟机制的资源协同分配方法。首先，对Globus和Legion的资源协同分配的方法进行了比较。然后，描述了一个基于Agent的网格资源管理结构。最后，设计了一个Agent联盟机制用于有效的资源协同分配。     

Resource allocation algorithms for virtualized service hosting platforms

Commodity clusters are used routinely for deploying service hosting platforms. Due to hardware and operation costs, clusters need to be shared among multiple services. Crucial for enabling such shared hosting platforms is virtual machine (VM) technology, which allows consolidation of hardware resources. A key challenge, however, is to make appropriate decisions when allocating hardware resources to service instances. In this work we propose a formulation of the resource allocation problem in shared hosting platforms for static workloads with servers that provide multiple types of resources. Our formulation supports a mix of best-effort and QoS scenarios, and, via a precisely defined objective function, promotes performance, fairness, and cluster utilization. Further, this formulation makes it possible to compute a bound on the optimal resource allocation. We propose several classes of resource allocation algorithms, which we evaluate in simulation. We are able to identify an algorithm that achieves average performance close to the optimal across many experimental scenarios. Furthermore, this algorithm runs in only a few seconds for large platforms and thus is usable in practice.     

A reinforcement learning approach to dynamic resource allocation

This paper presents a general framework for performing adaptive reconfiguration of a distributed system based on maximizing the long-term business value, defined as the discounted sum of all future rewards and penalties. The problem of dynamic resource allocation among multiple entities sharing a common set of resources is used as an example. A specific architecture (DRA-FRL) is presented, which uses the emerging methodology of reinforcement learning in conjunction with fuzzy rulebases to achieve the desired objective. This architecture can work in the context of existing resource allocation policies and learn the values of the states that the system encounters under these policies. Once the learning process begins to converge, the user can allow the DRA-FRL architecture to make some additional resource allocation decisions or override the ones suggested by the existing policies so as to improve the long-term business value of the system. The DRA-FRL architecture can also be deployed in an environment without any existing resource allocation policies. An implementation of the DRA-FRL architecture in Solaris 10 demonstrated a robust performance improvement in the problem of dynamically migrating CPUs and memory blocks between three resource partitions so as to match the stochastically changing workload in each partition, both in the presence and in the absence of resource migration costs.     

云环境下公平性优化的资源分配方法

针对云数据中心资源分配不均、效率不高、资源错位等问题,为了满足不同用户的需求,达到多种资源分配的公平性,实现资源的高效利用,提出了全局优势资源公平（GDRF）分配算法。GDRF算法采用多轮分配方式,即先通过用户已分配资源量确定分配资格,每轮再通过全局优势资源共享比和全局优势资源权重来确定具体的分配用户,分配过程充分考虑了资源的匹配情况,采用了max-min fairness思想的渐进填充方式,并且将多资源分配公平性统一度量模型运用到了算法中。实验基于一个Google集群数据模型与基于占优资源的多资源联合公平分配算法作了比较。实验结果表明,GDRF算法分配的虚拟机总量提高了12%,资源总利用率提高了0.5个百分点,公平评估值提高了约15%,并且该算法的资源组合分配的适应度较高,使得用户需求和供给更匹配。     

离散比特多播OFDM系统自适应资源分配算法

考虑具有多个多播组的OFDM系统中的资源分配问题,在保证各多播业务误比特率要求下,最大化系统频谱效率.针对实际系统中调制阶数为离散值的情况,首先将连续比特资源分配算法扩展到离散比特的情况,系统性能损失较大.为提高系统性能,提出一种基于多描述编码的离散比特资源分配算法,利用多描述编码灵活的解码特性,更有效地进行资源分配.仿真结果表明,所提基于多描述编码的离散比特资源分配算法,在不同多播组数目和每组用户数情况下,均可改善系统性能.     

OFDMA多维无线资源联合快速分配算法

OFDMA中多维无线资源的分配是无线通信的重要研究课题之一.通过分部分配的方法,OFDMA系统中的多维无线资源分配算法能够有效降低无线资源分配的时间复杂度,实现资源的快速分配.在此基础上,基于流平衡的资源分配算法综合考虑了用户加权业务速率以及全体用户的归一化子载波容量,能够对多维无线资源联合进行分配.仿真表明,与传统的无线资源分配算法M-LWDF相比,流平衡分部分配算法能更好地适用于多维无线资源系统,在相同的时间复杂度下能够拥有更大的频谱利用率,达到良好的系统性能.     

Feedback control for multi-resource usage of virtualised database server

Virtualisation and cloud computing have recently received significant attention. Resource allocation and control of multiple resource usages among virtual machines in virtualised data centres remains an open problem. Therefore, in this paper, our focus is to control CPU (central processing unit) usage and memory consumption of a virtual database machine in a data centre under a time-varying heavy workload. In addition to existing work, we attempt to control multiple outputs, such as the CPU usage and memory consumption of a virtualised database server (DBVM), via changing multiple server parameters, such as the CPU allocation and memory allocation, in real time. We indicated that a virtualised database server might be modelled as a linear time-unvarying system. We obtained and compared both MIMO (multi input–multi output) and multiple SISO (single input–single output) models of that system. We designed multiple SISO feedback controllers to achieve desired CPU usages and memory consumptions under workload.     

Online optimization for scheduling preemptable tasks on IaaS cloud systems

In Infrastructure-as-a-Service (IaaS) cloud computing, computational resources are provided to remote users in the form of leases. For a cloud user, he/she can request multiple cloud services simultaneously. In this case, parallel processing in the cloud system can improve the performance. When applying parallel processing in cloud computing, it is necessary to implement a mechanism to allocate resource and schedule the execution order of tasks. Furthermore, a resource optimization mechanism with preemptable task execution can increase the utilization of clouds. In this paper, we propose two online dynamic resource allocation algorithms for the IaaS cloud system with preemptable tasks. Our algorithms adjust the resource allocation dynamically based on the updated information of the actual task executions. And the experimental results show that our algorithms can significantly improve the performance in the situation where resource contention is fierce.     

Modeling and analysis of air campaign resource allocation: a spatio-temporal decomposition approach

In this paper, we address the modeling and analysis issues associated with a generic theater level campaign where two adversaries pit their military resources against each other over a sequence of multiple engagements. In particular, we consider the scenario of an air raid campaign where one adversary uses suppression of enemy air defense (SEAD) aircraft and bombers (BMBs) against the other adversary's invading ground troops (GTs) that are defended by their mobile air defense (AD) units. The original problem is decomposed into a temporal and a spatial resource allocation problem. The temporal resource allocation problem is formulated and solved in a game-theoretical framework as a multiple resource interaction problem with linear attrition functions. The spatial resource allocation problem is posed as a risk minimization problem in which the optimal corridor of ingress and optimal movement of the GTs and AD units are decided by the adversaries. These two solutions are integrated using an aggregation/deaggregation approach to evaluate resource strengths and distribute losses. Several simulation experiments were carried out to demonstrate the main ideas.     

统计推荐模型中的异构数据源资源配置

信息系统中的统计推荐模型需要获取、分析和汇总多个来源的数据。这些多源异构的数据集在特征和价值方面可能存在显著差异,因而影响模型性能。为了提升统计推荐模型的整体性能,这项研究工作采用凸优化理论和方法,解决了统计推荐模型中异构数据源的最优资源配置问题。在不同的数据源资源配置下,该工作对比了同一推荐模型的性能变化。实验结果表明该工作提出的资源分配算法在NDCG（normalized discounted cumulative gain）和召回率这两个推荐系统主要评价指标上有效地提升了模型性能。这项工作的结论是：针对多个异构的数据源,适当的资源划分和分配策略可以显著影响推荐模型的整体性能。     

Game Theoretic subcarrier and power allocation for wireless OFDMA networks

This paper proposes a bargaining game theoretic resource(including the subcarrier and the power) allocation scheme for wireless orthogonal frequency division multiple access(OFDMA) networks.We define a wireless user s payoff as a function of the achieved data-rate.The fairness resource allocation problem can then be modeled as a cooperative bargaining game.The objective of the game is to maximize the aggregate payoffs for the users.To search for the Nash bargaining solution(NBS) of the game,a suboptimal subcarrier allocation is performed by assuming an equal power allocation.Thereafter,an optimal power allocation is performed to maximize the sum payoff for the users.By comparing with the max-rate and the max-min algorithms,simulation results show that the proposed game could achieve a good tradeoff between the user fairness and the overall system performance.