基于滚动优化的虚拟云中实时任务节能调度方法

目前,节能已成为云数据中心的研究热点.建设节能的云数据中心不仅可以减少用电消耗,而且可以提高系统的可靠性.现有的云中心节能调度算法缺乏在任务调度级别的考虑,使得任务执行效果受到较大影响.为此,首先给出了一种基于滚动优化的实时任务调度器结构,然后详细分析和构建了任务能量消耗模型.在此基础上提出了一种实时非周期任务节能调度算法EARH(energy-aware scheduling algorithm).EARH采用的滚动优化策略能够被拓展并集成其他节能调度算法.此外,提出了资源动态增加与缩减策略,用于在系统可调度性与节能两方面进行权衡.最后,通过大量的模拟实验验证了EARH的性能.与其他3种基准算法相比,其实验结果表明,EARH的调度质量优于其他算法,可有效提高系统性能.     

云计算中资源延迟感知的实时任务调度方法

绿色云计算已经成为一个研究焦点,动态整合虚拟机和关闭空闲主机是极具潜力的途径可降低云计算数据中心的能耗.当云平台的负载迅速增加时,系统需要启动更多的主机和创建更多的虚拟机来扩展可用资源.然而,启动主机和创建虚拟机需要一定的时间开销,使得紧急任务难以及时开始,从而延误了截止期.为了解决以上问题,首先提出具有机器启动时间感知的虚拟机扩展策略,以缓解机器启动时间冲击实时任务的时效性要求.基于该策略,设计算法STARS来调度实时任务和资源,以在保障任务时效性与节能2方面进行权横.最后,使用Google的负载数据进行模拟实验,比较算法STARS与其他2个算法的性能.实验结果表明,在保障任务时效性、节能和资源利用率方面,算法STARS优于对比算法.     

Iterative-Improvement-Based Heuristics for Adaptive Scheduling of Tasks Sharing Files on Heterogeneous Master-Slave Environments

The scheduling of independent but file-sharing tasks on heterogeneous master-slave platforms has recently found important applications in Grid environments. The scheduling heuristics recently proposed for this problem are all constructive in nature and based on a common greedy criterion which depends on the momentary completion time values of the tasks. We show that this greedy decision criterion has shortcomings in exploiting the file-sharing interaction among tasks since completion time values are inadequate to extract the global view of this interaction. We propose a three-phase scheduling approach which involves initial task assignment, refinement, and execution ordering phases. For the refinement phase, we model the target application as a hypergraph and, with an elegant hypergraph-partitioning-like formulation, we propose using iterative-improvement-based heuristics for refining the task assignments according to two novel objective functions. Unlike the turnaround time, which is the actual schedule cost, the smoothness of proposed objective functions enables the use of iterative-improvement-based heuristics successfully since their effectiveness and efficiency depend on the smoothness of the objective function. Experimental results on a wide range of synthetically generated heterogeneous master-slave frameworks show that the proposed three-phase scheduling approach performs much better than the greedy constructive approach.     

基于辅助队列的硬实时混合任务节能调度算法

节能调度是当今实时系统研究的一个重要领域,其中混合实时任务节能调度技术研究刚刚起步.OLDVS算法是非常简洁的硬实时系统在线节能调度算法,但存在以下不足:不适应任务执行的动态变化,不能有效利用动态松弛时间,过于保守以致节能效果并不理想.据此,提出一种新的基于辅助队列的硬实时混合任务节能调度算法(OLDVS-AQ).通过引入一个额外的数据结构即辅助队列(Assisted Queue,AQ)来计算任务的最大完成时间,能够更有效地利用动态松弛时间进一步降低能耗.证明了该算法的可调度性,仿真实验结果表明,OLDVS-AQ算法始终优于OLDVS算法.平均提高约10%的节能效果.     

任务具有链约束的平行机调度问题

研究任务间具有链约束的平行机调度问题,目标是在满足任务间链约束的条件下任务的总完工时间最小,这类问题是NP-难的.通过对问题的分析,对于一般情况给出了最优解的必要条件,对于特殊情况给出了问题的最优解算法.     

虚拟集群上面向功耗的形式化的VM调度策略

针对虚拟化集群上日益严重的功耗问题,在定义集群、物理机、操作点、能耗、任务集、虚拟机等概念的基础上,提出了形式化的虚拟机(Virtual Machine,VM)调度策略,并有针对性地利用动态电压和频率调整(Dynamic Voltage and Frequency Scaling,DVFS)技术对普通算法进行了扩展改进,提出一个功耗敏感的VM调度算法。首先,利用FFD(First-Fit Decreasing)算法解决虚拟机的初始布局问题,然后尽可能地将虚拟机部署在低电压的物理机上,此外加入性能感知策略,尽量将物理机的电压调节至刚好满足虚拟机性能需求,以避免"奢侈"能耗的浪费,同时增加了虚拟机调度失败时灵活调整电压的规则,以减少错误先验知识对虚拟机后续部署的误导。在理想模型中的仿真实验表明:与现有部署算法相比,本算法具有更好的节能效应,以及相对不大的性能损失。     

双层虚拟化云架构下任务调度能耗优化算法

虚拟机上部署容器的双层虚拟化云架构在云数据中心中的使用越来越广泛.为了解决该架构下云数据中心的能耗问题,提出了一种工作流任务调度算法TUMS-RTC.针对有截止时间约束的并行工作流,算法将调度过程划分为时间利用率最大化调度和运行时间压缩两个阶段.时间利用率最大化调度通过充分使用给定的时间范围减少完成工作流所需的虚拟机和...     

QoS约束云环境下的工作流能效调度算法

云环境可以为大规模工作流的执行提供高效、可靠的运行环境,但工作流执行时带来的高能耗不仅会增加云资源提供方的经济成本,还会影响云系统的可靠性,并对环境产生不利影响。为了在满足用户截止时间QoS需求的同时降低云环境中工作流调度的执行能耗,提出一种工作流能效调度算法QCWES。该算法将工作流的能效调度方案求解划分为3个阶段:截止时间重分配、任务调度选择排序以及基于DVFS的最佳资源选择。截止时间重分配阶段旨在将用户定义的全局工作流截止时间在各个任务间进行重分配,任务调度选择排序阶段旨在通过自顶向下的任务分级方式得到任务调度序列;基于DVFS的最佳资源选择阶段旨在为每个任务选择带有合适电压/频率等级的最优目标资源,在满足任务的子截止时间的前提下使总体能耗达到最小。通过随机工作流和基于高斯消元法的现实工作流结构,对算法的性能进行仿真实验分析。结果表明,所提算法可以在满足截止时间约束下降低工作流的执行能耗,实现用户方的QoS需求与资源方的能耗间的均衡。     

一种在线节能实时调度算法

随着多处理器系统规模的不断扩大,如何节能成为一个亟待解决的重要问题。为此,基于多处理器系统提出一种针对随机任务的在线节能实时调度算法。使用统计方法,根据已有任务的到达时间和计算量估计新任务在空闲处理器上执行的电压/频率,使还未到达的任务能够满足截止期限并有效节能。在考虑单个处理器上执行的任务时,计算执行这些任务所需的平均电压/频率,使所有任务的执行速度尽量均衡,当某些任务不能满足截止期限要求时,则调高未执行任务的电压/频率。实验结果表明,与EDF,HVEA,MEG和ME-MC算法相比,该算法在满足截止期限和节能方面具有明显的优势。     

Scheduling Independent Multiprocessor Tasks

We study the problem of scheduling a set of n independent multiprocessor tasks with prespecified processor allocations on a fixed number of processors. We propose a linear time algorithm that finds a schedule of minimum makespan in the preemptive model, and a linear time approximation algorithm that finds a schedule of makespan within a factor of (1+\eps) of optimal in the non-preemptive model. We extend our results by obtaining a polynomial time approximation scheme for the parallel processors variant of the multiprocessor task model.     

Scheduling Independent Multiprocessor Tasks

Abstract. We study the problem of scheduling a set of n independent multiprocessor tasks with prespecified processor allocations on a fixed number of processors. We propose a linear time algorithm that finds a schedule of minimum makespan in the preemptive model, and a linear time approximation algorithm that finds a schedule of makespan within a factor of (1+\eps) of optimal in the non-preemptive model. We extend our results by obtaining a polynomial time approximation scheme for the parallel processors variant of the multiprocessor task model.     

基于动态抢占阈值的实时调度

具有抢占阈值的调度算法集非抢占调度和纯抢占调度的特点,既减少了由于过多的随意抢占造成的CPU资源浪费,又保证了一定的任务截止期错失率及CPU资源利用率。已有的工作基本集中于讨论任务集完全给定,任务数、任务的优先级及任务的抢占阈值在调度前已完全确定,而且要求不同的任务具有不同的优先级,提出的具有抢占阈值的调度算法,完全放松了对这些条件的限制,即任务的个数不确定,任务的优先级及其抢占阈值在调度过程中可以动态地变化。最后以常用的LSF调度策略为例,结合动态的抢占阈值进行仿真,仿真结果表明,对于不确定的任务集、任务优先级和抢占阈值,利用具有抢占阈值的动态调度算法,降低了任务截止期错失率、提高了CPU的有效使用率。     

Scheduling Tasks with Resource Requirements in Hard Real-Time Systems

This paper describes a heuristic approach for solving the problem of dynamically scheduling tasks in a real-time system where tasks have deadlines and general resource requirements. The crux of our approach lies in the heuristic function used to select the task to be scheduled next. The heuristic function is composed of three weighted factors. These factors explicitly consider information about real-time constraints of tasks and their utilization of resources. Simulation studies show that the weights for the various factors in the heuristic function have to be fine-tuned in order to obtain a degree of success in the range of 75-88 percent of that obtained via exhaustive search. However, modifying the approach to use limited backtracking improves the degree of success substantially to as high as 99.5 percent. This improvement is observed even when the initial set of weights are not tailored for a particular set of tasks. Simulation studies also show that in most cases the schedule determined by the heuristic algorithm is optimal or close to optimal.     

An Energy-Aware Heuristic Scheduling for Data-Intensive Workflows in Virtualized Datacenters

With the development of cloud computing, more and more data-intensive workflows have been deployed on virtualized datacenters. As a result, the energy spent on massive data accessing grows rapidly. In this paper, an energy-aware scheduling algorithm is proposed, which introduces a novel heuristic called Minimal Data-Accessing Energy Path for scheduling data-intensive workflows aiming to reduce the energy consumption of intensive data accessing. Extensive experiments based on both synthetical and real workloads are conducted to investigate the effectiveness and performance of the proposed scheduling approach. The experimental results show that the proposed heuristic scheduling can significantly reduce the energy consumption of storing/retrieving intermediate data generated during the execution of data-intensive workflow. In addition, it exhibits better robustness than existing algorithms when cloud systems are in presence of I/O- intensive workloads.     

Dynamic Real-time Scheduling of Firm Periodic Tasks with Hard and Soft Aperiodic Tasks

In this paper, we address the problem of the dynamic scheduling of skippable periodic task sets (i.e., period tasks allowing occasional skips of instances), together with aperiodic tasks. Scheduling of tasks is handled thanks to the merging of two existing approaches: the Skip-Over task model and the EDL (Earliest Deadline as Late as possible) aperiodic task server. The objective is to provide two on-line scheduling algorithms, namely EDL-RTO and EDL-BWP, in order to minimize the average response time of soft aperiodic requests, while ensuring that the QoS (Quality of Service) of periodic tasks will never be less than a specified bound. We also extend our results to the acceptance of sporadic tasks (i.e., aperiodic tasks with deadlines). We show that these novel scheduling algorithms have better performance compared to related algorithms regarding aperiodic response time and acceptance ratio. Audrey Marchand guaduated in Computer Engineering at the Ecole polytechnique of the University of Nantes (France), in 2002. She is currently a PhD student at the University of Nantes. Her research interests include real-time scheduling theory, aperiodic service mechanisms, quality of service guarantees in soft real-time systems, and Linux-based real-time operating systems and applications. Maryline Chetto received the degree of Docteur de 3ème cycle in control engineering and the degree of Habilitée à Diriger des Recherches in Computer Science from the University of Nantes, France, in 1984 and 1993, respectively. From 1984 to 1985, she held the position of Assistant professor of Computer Science at the University of Rennes, while her research was with the Institut de Recherche en Informatique et Systèmes Aléatoires, Rennes. In 1986, she returned to Nantes and is currently a professor with the Institute of Technology of the University of Nantes. She is conducting her research at IRCCyN. Her main research interests include scheduling and fault-tolerance technologies for real-time applications. She has published more than 60 journal articles and conference papers in the area of real-time operating systems. She is the leader of a French national R&D project, namely Cleopatre, supported by the French government, which aims to provide free open source real-time solutions.     

成本约束的网格工作流时间优化方法

针对成本约束有向无环图DAG(directed acyclic graph)表示的网格工作流完工时间最小化问题,提出两个基于优先级规则的迭代启发算法.算法利用并行活动特征定义正向分层和逆向分层两个概念,将其分别引入最大收益规则MP(maximum profit),得到正分层最大收益规则MPTL(maximum profit with top level) 和逆分层最大收益规则MPBL(maximum profit with bottom level).两规则每次迭代尽量以完工时间的最小增加换取总费用的最大降低,逐步将分层初始解构造为满足成本约束的可行解.模拟结果表明,两规则在获得较少迭代次数和运行时间的同时,能显著改进MP规则的平均性能,且MPBL优于MPTL.     

Heuristics for the Unrelated Parallel Machine Scheduling Problem with Setup Times

The problem addressed in this paper is the non-preemptive unrelated parallel machine scheduling problem with the objective of minimizing the makespan. Machine-dependent and job sequence-dependent setup times are considered, all jobs are available at time zero, and all times are deterministic. This is a NP-hard problem and in this paper, optimal solutions are found for small problems only. For larger problems, a new meta-heuristic, Meta-RaPS, is introduced and its performance is evaluated by comparing its solutions to the solutions of an existing heuristic for the same problem. The results show that Meta-RaPS found all optimal solutions for the small problems and outperformed the solutions obtained by the existing heuristic for larger problems.     

Data-Oriented Scheduling with Dynamic-Clustering Fault-Tolerant Technique for Scientific Workflows in Clouds

Programming and Computer Software - Cloud computing is one of the most prominent parallel and distributed computing paradigm. It is used for providing solution to a huge number of scientific and...     

Learning Heuristics for the Superblock Instruction Scheduling Problem

Modern processors have multiple pipelined functional units and can issue more than one instruction per clock cycle. This places a burden on the compiler to schedule the instructions to take maximum advantage of the underlying hardware. Superblocks—a straight-line sequence of code with a single entry point and multiple possible exit points—are a commonly used scheduling region within compilers. Superblock scheduling is NP-complete, and is done suboptimally in production compilers using a greedy algorithm coupled with a heuristic. The heuristic is usually handcrafted, a potentially time-consuming process. In this paper, we show that supervised machine learning techniques can be used to semiautomate the construction of heuristics for superblock scheduling. In our approach, labeled training data were produced using an optimal superblock scheduler. A decision tree learning algorithm was then used to induce a heuristic from the training data. The automatically constructed decision tree heuristic was compared against the best previously proposed, handcrafted heuristics for superblock scheduling on the SPEC 2000 and MediaBench benchmark suites. On these benchmark suites, the decision tree heuristic reduced the number of superblocks that were not optimally scheduled by up to 38 percent, and led to improved performance on some architectural models and competitive performance on others.