基于异构感知静态调度与动态迁移的异构多核调度机制

异构多核处理器体系结构可以有效减少功效开销,是处理器发展的趋势,负载不平衡问题会造成处理器执行的不稳定。提出一种基于异构感知的静态调度和动态线程迁移相结合的异构多核调度机制,解决了不同核之间的负载平衡问题,提高了吞吐量。仿真实验通过将此调度机制与静态调度策略（SS）比较,表明该机制提高了异构多核处理器的性能并保证了执行过程的稳定性。     

基于多核处理器的K线程低能耗的任务调度优化算法

针对具有独立DVFS的多核处理器系统,提出了一种K线程低能耗模型的并行任务调度优化算法(Tasks Optimization based on Energy-Effectiveness Model,TO-EEM)。与传统的并行任务节能调度相比,该算法的主要目标是不仅通过降低处理器频率来减少处理器瞬时功耗,而且结合并行任务间的同步互斥所造成的线程阻塞情况,合理分配线程资源来减少线程同步时间,优化并行性能;保证任务在一定的并行加速比性能前提下,提高资源利用率,减少能耗,达到程序能耗和性能之间的折衷。文中进行了大量模拟实验,结果证明提出的任务优化模型算法节能效果明显,能有效降低处理器的功耗,并始终保持线性加速比。     

多核系统中基于动态电压频率调节的实时节能调度研究

在实时嵌入式领域,特别是无线移动和便携式计算领域,能耗是首要考虑的因素,这也是多核处理器尚未在嵌入式领域全面展开应用的首要因素。目前针对多核系统的实时应用,基于动态电压频率调节(DVFS)的实时节能调度技术研究得较少,还有许多问题亟待解决。本文介绍了多核系统中动态电压频率调节技术,分析讨论了当前多核系统中实时调度研究进展,主要针对同构多核、异构多核、并行任务模型和弱硬实时模型等方面,深入探讨了多核系统中基于DVFS的实时节能调度。本文结合多核系统、电压频率动态调节节能和实时调度,探索了多核系统中的实时节能调度,奠定了理论和技术基础,具有重大的理论意义和现实应用价值。     

面向动态异构多核处理器的公平调度算法

动态异构多核处理器的处理器核可动态调整的特征给操作系统调度算法带来了新的机遇和挑战.利用处理器核动态可调整的特征能更好地适应不同任务的运行需求,带来巨大的性能优化空间.然而也带来新的代价和更复杂的公平性的计算.为了解决面向动态异构多核处理器结构上的公平性调度问题,提出了一个基于集中式运行队列的调度模型,以降低调度算法在动态处理器核变化所带来的维护开销.并重新思考在动态异构处理器结构下公平性的定义,基于原有CFS调度算法提出新的HFS调度算法.HFS调度算法不仅能简单而有效地利用动态异构多核处理器的性能优势,而且能提供在动态异构多核处理器上的公平性调度.通过模拟SCMP,ACMP,DHCMP平台,证明了提出的HFS调度算法能够很好地发挥DHCMP结构的性能特征,比运行目前主流调度算法的SCMP和ACMP结构提升10.55%的用户级性能(ANTT),14.24%的系统吞吐率(WSU).     

异构多核上支持OpenMP3.0的自适应任务粒度策略

任务粒度是决定任务并行程序性能的关键因素,鉴于不同应用其最优的任务粒度可能不同,提出一种异构多核Cell处理器上支持OpenMP3.0的自适应任务粒度策略.该策略首先广度生成任务,直到所有的线程达到饱和,之后若某个线程执行完自身任务而处于空闲状态时,通过回溯到忙碌线程的任务树中最早可以派生任务的结点处生成新任务,以供空闲线程窃取执行.该策略不仅保证生成的任务粒度最大化,并且有效地解决了负载不均衡问题.实验在一个Cell处理器上进行,结果表明与顺序执行速度相比,自适应任务粒度策略达到了4.1到7.2的加速比,并且该策略优于现有的Tascell和AdaptiveTC方案,同时对于绝大部分应用表现出了良好的可扩展行.     

S-Bridge:性能非对称多核处理器下负载均衡代理机制

近年来,在移动计算环境中,异构多核处理器已经逐渐成为主流.与传统同构的处理器设计相比,此类异构多核处理器以更低的功耗成本满足设备的计算需求.但是异构环境下CPU核之间的微架构差异,也为操作系统中的一些基本方法提出了新的挑战.面向性能非对称异构多核环境下调度的负载均衡问题,从系统层面提出了一种负载均衡机制S-Bridge,可以减少处理器微架构差异以及任务执行需求差异对传统负载均衡带来的影响.S-Bridge的主要贡献是从系统层提供了通用的、适配异构性的负载均衡相关接口,使任意调度器都能方便地与异构多核处理器系统进行适配.基于CFS和HMP调度器在ARM平台上进行实验,同时在X86平台上进行S-Bridge通用性的验证,结果表明：S-Bridge可以支持不同真实平台和内核版本的快速实现,平均性能提升超过15%,部分情况下可达65%.     

基于big.LITTLE架构的多目标功耗自适应控制方法

针对移动计算系统功耗约束条件时常变动,以及动态电压频率调节无法有效克服静态功耗导致的能量损失等问题,提出一种多目标功耗自适应控制方法。根据实时功耗约束制定调核策略,确定处理器核类型及数量,结合操作系统线程亲和性、进程迁移与处理器热插拔完成处理器核的开启、关闭及负荷管理,实现功耗自适应。在典型多核应用MapReduce模型Phoenix与可变形部件模型上的实验结果表明,该方法能够按需调度核类型及数量来完成计算任务,与传统功率恒定系统相比,执行时间与能耗平均减少60.91%和48.54%,有效提高目标系统能效。     

基于粒子群算法的多核处理器线程调度研究

为有效解决多核处理器的线程调度问题,提出了一种基于粒子群算法框架上的线程调度算法.该算法依据设计的调度模型,在线程DAG图上通过复制不在同一处理器上且存在相关性的线程,生成相互独立的子DAG图,并采用改进的粒子群优化算法对其进行合理调度,由此提高线程调度效率.仿真实现了该算法,并通过实验数据验证了该算法的优越性.     

异构多核处理器的任务分配及能耗的研究*

异构多核处理器采用不同的任务分配与调度算法,会导致不同的时间消耗与能量消耗,采用合适的任务分配与调度算法能节省较多的能耗。目前普遍认为最有发展前途的任务分配与调度技术是先用启发式方法进行分组,然后再用遗传算法进行调度。在改进任务分组后,又首次提出了用遗传算法解决能耗问题。实验结果表明在实时要求不高的情况下,能以较小的时间代价来节省较多的能耗。     

高实时性异构多核处理器任务调度算法

在异构多核处理器条件下,Min-Min算法调度性能较好但在系统实时响应方面存在不足。最小空闲时间优先调度算法(LSF)、最早截止时间优先调度算法(EDF)和最大价值优先调度算法(HVF)虽然在系统任务调度响应实时性方面表现优异,但却不适用于异构多核处理器环境。为此,提出一种高实时性任务调度算法HRSA。在Min-Min调度算法的基础上融合LSF,EDF,HVF算法的调度策略,将任务能耗、任务完成价值和任务响应比相结合,在实现异构多核处理器任务动态调度的同时缩短系统对高实时性任务的响应时间。实验结果表明,相对于EDF算法和Min-Min算法,HRSA算法消耗单位能量所带来的价值较高,对高实时性任务处理的响应时间较短。     

Energy efficient scheduling of real-time tasks on multi-core processors with voltage islands

This paper studies energy efficient scheduling of periodic real-time tasks on multi-core processors with voltage islands, in which cores are partitioned into multiple blocks (termed voltage islands) and each block has its own power source to supply voltage. Cores in the same block always operate at the same voltage level, but can be adjusted by using Dynamic Voltage and Frequency Scaling (DVFS). We propose a Voltage Island Largest Capacity First (VILCF) algorithm for energy efficient scheduling of periodic real-time tasks on multi-core processors. It achieves better energy efficiency by fully utilizing the remaining capacity of an island before turning on more islands or increasing the voltage level of the current active islands. We provide detailed theoretical analysis of the approximation ratio of the proposed VILCF algorithm in terms of energy efficiency. In addition, our experimental results show that VILCF significantly outperforms the existing algorithms when there are multiple cores in a voltage island.     

非对称多核处理器上的操作系统集成调度

相对于对称多核处理器,非对称多核处理器具有更高的效能,将成为未来并行操作系统中的主流体系结构.对于非对称多核处理器上操作系统的并行任务调度问题,现有的研究假设所有核心频率恒定,缺乏理论分析,也没有考虑算法的效能和通用性.针对该问题,该文首先建立非线性规划模型,分析得出全面考虑并行任务同步特性、核心非对称性以及核心负载的调度原则.然后,基于调度原则提出一个集成调度算法,该算法通过集成线程调度和动态电压频率调整来提高效能,并通过参数调整机制实现了算法的通用性.提出的算法是第一个在非对称多核处理器上结合线程调度和动态电压频率调整的调度算法.实际平台上的实验表明:该算法可适用于多种环境,且效能比其他同类算法高24%～50%.     

An Energy-Efficient Task Scheduling Algorithm in DVFS-enabled Cloud Environment

The growth of energy consumption has been explosive in current data centers, super computers, and public cloud systems. This explosion has led to greater advocacy of green computing, and many efforts and works focus on the task scheduling in order to reduce energy dissipation. In order to obtain more energy reduction as well as maintain the quality of service by meeting the deadlines, this paper proposes a DVFS-enabled Energy-efficient Workflow Task Scheduling algorithm: DEWTS. Through merging the relatively inefficient processors by reclaiming the slack time, DEWTS can leverage the useful slack time recurrently after severs are merged. DEWTS firstly calculates the initial scheduling order of all tasks, and obtains the whole makespan and deadline based on Heterogeneous-Earliest-Finish-Time (HEFT) algorithm. Through resorting the processors with their running task number and energy utilization, the underutilized processors can be merged by closing the last node and redistributing the assigned tasks on it. Finally, in the task slacking phase, the tasks can be distributed in the idle slots under a lower voltage and frequency using DVFS technique, without violating the dependency constraints and increasing the slacked makespan. Based on the amount of randomly generated DAGs workflows, the experimental results show that DEWTS can reduce the total power consumption by up to 46.5 % for various parallel applications as well as balance the scheduling performance.     

A green energy-efficient scheduling algorithm using the DVFS technique for cloud datacenters

Information and communication technology (ICT) has a profound impact on environment because of its large amount of CO₂ emissions. In the past years, the research field of “green” and low power consumption networking infrastructures is of great importance for both service/network providers and equipment manufacturers. An emerging technology called Cloud computing can increase the utilization and efficiency of hardware equipment. The job scheduler is needed by a cloud datacenter to arrange resources for executing jobs. In this paper, we propose a scheduling algorithm for the cloud datacenter with a dynamic voltage frequency scaling technique. Our scheduling algorithm can efficiently increase resource utilization; hence, it can decrease the energy consumption for executing jobs. Experimental results show that our scheme can reduce more energy consumption than other schemes do. The performance of executing jobs is not sacrificed in our scheme. We provide a green energy-efficient scheduling algorithm using the DVFS technique for Cloud computing datacenters.     

Scheduling with dynamic voltage/speed adjustment using slack reclamation in multiprocessor real-time systems

The high power consumption of modern processors becomes a major concern because it leads to decreased mission duration (for battery-operated systems), increased heat dissipation, and decreased reliability. While many techniques have been proposed to reduce power consumption for uniprocessor systems, there has been considerably less work on multiprocessor systems. In this paper, based on the concept of slack sharing among processors, we propose two novel power-aware scheduling algorithms for task sets with and without precedence constraints executing on multiprocessor systems. These scheduling techniques reclaim the time unused by a task to reduce the execution speed of future tasks and, thus, reduce the total energy consumption of the system. We also study the effect of discrete voltage/speed levels on the energy savings for multiprocessor systems and propose a new scheme of slack reservation to incorporate voltage/speed adjustment overhead in the scheduling algorithms. Simulation and trace-based results indicate that our algorithms achieve substantial energy savings on systems with variable voltage processors. Moreover, processors with a few discrete voltage/speed levels obtain nearly the same energy savings as processors with continuous voltage/speed, and the effect of voltage/speed adjustment overhead on the energy savings is relatively small.     

一种基于同构多核处理器的动态节能调度算法

在节能计算系统中节能调度是一个重要的研究方向。目前针对节能调度的研究中,多从动态电压频率调节（DVFS）的角度进行。但是随着芯片制造工艺的改进,处理器核心电压越来越低,所能调节的电压范围越来越小,通过DVFS技术实现节能遇到了理论上的瓶颈。提出了一种基于同构多核处理器的动态节能调度算法。在系统负载较轻的情况下,通过将系统任务分配到较少的处理核心上而使其他处理核心处于休眠状态来达到动态节能的效果。实验表明该算法具有较好的节能效果。     

Effectiveness Analysis of DVFS and DPM in Mobile Devices

The demand for high-performance embedded processors in multimedia mobile electronics is growing and their power consumption thus increasingly threatens battery lifetime.It is usually believed that the dynamic voltage and frequency scaling (DVFS) feature saves significant energy by changing the performance levels of processors to match the performance demands of applications on the fly.However,because the energy efficiency of embedded processors is rapidly improving,the effectiveness of DVFS is expected to change.In this paper,we analyze the benefit of DVFS in state-of-the-art mobile embedded platforms in comparison to those in servers or PCs.To obtain a clearer view of the relationship between power and performance,we develop a measurement methodology that can synchronize time series for power consumption with those for processor utilization.The results show that DVFS hardly improves the energy efficiency of mobile multimedia electronics,and can even significantly worsen energy efficiency and performance in some cases.According to this observation,we suggest that power management for mobile electronics should concentrate on adaptive and intelligent power management for peripheral devices.As a preliminary design,we implement an adaptive network interface card (NIC) speed control that reduces power consumption by 10% when NIC is not heavily used.Our results provide valuable insights into the design of power management schemes for future mobile embedded systems.     

一种云工作流任务调度能效优化算法

工作流任务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,而且会降低云系统的可靠性。为了满足截止时间的同时,降低工作流执行能耗,提出一种工作流能效调度算法CWEES。算法将能效优化调度划分为三个阶段：初始任务映射、处理器资源合并和任务松驰。初始任务映射旨在通过任务自底向上分级排序得到任务调度初始序列,处理器资源合并旨在通过重用松驰时间合并相对低效率的处理器,降低资源使用数量,任务松驰旨在为每个任务重新选择带有合适电压/频率等级的最优目标资源,在不违背任务顺序和截止时间约束前提下降低工作流执行总能耗。通过随机工作任务模型对算法的性能进行了仿真实验分析。结果表明,CWEES算法不仅资源利用率更高,而且可以在满足截止时间约束下降低工作流执行能耗,实现执行效率与能耗的均衡。     

Exploiting dynamic micro-architecture usage in gate sizing

Modern high performance microprocessors incorporate an abundance of replicated structural components. Many of these components often experience substantially lower utilization while executing a diverse pool of applications. To recover energy efficiency from the lower utilization, system architects resort to dynamic voltage frequency scaling (DVFS). In this paper, we demonstrate that dynamic adaptations using DVFS are markedly energy inefficient than techniques that design circuits ground up for lower performance. We propose a novel microarchitecture aware gate sizing and threshold voltage assignment algorithm to mitigate this current limitation. Our technique is the first of its kind that exploits architectural slack in gate sizing, and leverages on-chip redundancy and slack. We evaluate this circuit-architectural co-optimization framework in a superscalar processor by combining standard cell based gate sizing flows with state-of-the-art architectural simulation. Our results show 17-46% improvement in the datapath energy efficiency over traditional circuit designs incorporating DVFS schemes.