分布式实时系统中的预测调度算法

对于分布式实时系统中的周期性任务,人们提出了一系列静态分配调度算法,有效地解决了各种特定条件下的任务分配和调度问题.这些算法的主要特点是,它们均要求被调度任务的特征参数为已知条件.然而在很多实时系统中,周期性任务的运行时间或任务数量常常是一些具有一定规律的随机过程,因而上述静态算法的效能将受到限制.在分析了特定应用背景中的处理流程之后,抽象得到两类随机任务模型,针对这两类模型介绍了在分布式实时系统中已经得到应用的静态分配调度算法SAA(static allocation algorithms),进而提出了多任务分配调度的预测算法PAA(predicting allocation algorithm).它根据周期性任务执行时间或子任务数量的统计特性,实现任务参量的合理预测和多任务的动态调度,以提高系统的实时性能.仿真结果表明,对于两类任务模型,PAA算法与SAA算法相比,在任务完成时间、负载均衡度、系统响应时间及任务夭折率等多方面均有显著改善.     

分布式实时系统的容错调度算法

提出了两种分布式实时容错调度算法：副版本后调度算法（ＢＫＣＬ）及无容错需求后调度算法（ＮＦＲＬ）,并研究了算法的时间复杂度,这两种容雕工算法能同时调度具有容错需求的实时任务和无容错需求的实时任务,ＢＫＣＬ和ＮＦＲＬ所产生的调度可保证：在分布式系统中一个节点机失效的情况下,具有容错需求的实时任务仍然可在截止时间内完成,在描述了两个实时容错调度算法之后,分别证明了这两个算法的容错调度正确性。接着,阐述     

实时系统中任务可调度性研究

讨论了有关实时调度理论,对实时系统采用了时间Ｐｅｔｒｉ Ｎｅｔ方法来建模,并运用实时调度理论对系统中的任务可调度性能进行了全面分析,给出了详细的分析步骤。     

异构分布式实时系统的容错优化调度算法

针对分布式实时系统,在分析了单处理调度算法的基础上,结合版本复制技术和首次适应方法,给出了一种容错调度算法。分析了算法的可调度性,给出任务的可调度性条件。在满足任务容错可调度的情况下,以提高处理器的利用率为目标,对基版本时限进行了优化,给出了基版本优化时限的求取算法。仿真结果表明,本文算法将可以得到比FTEDFFF和FTRMFF更高的处理器利用率。     

基于对象的分布式实时系统调度模型研究

为了解决分布式实时系统有关分配和调度等问题，给出并用形式化方法描述了一种基于对象分布式实时系统调度的通用模型。该模型包括表示时限的绝对时间约束，表示周期属性的周期约束，表示各种前趋关系和同步要求的相对时间约束以及保证资源使用一致性的一致性约束，此外该模型克服了以往模型不能在应用系统的逻辑和功能部件上描述系统实时的约束的不足，允许从方法和活动上描述所需的约束，降低了单一约束描述的繁杂程度，为了能够使用现有调度算法进行任务调度，讨论了约束转换的问题，给出了高层约束到底层约束的转换规则和相应的转换算法。     

实时系统调度算法综述

在多道程序环境下,主存中有多个进程,其数目往往多于处理机数目。操作系统通过处理机调度程序,按照某种调度算法动态地把处理机分配给就绪队列中的一个进程,使之执行。处理机是重要的计算机资源,提高处理机的利用率及改善系统性能（吞吐量、响应时间）,很大程度上取决于处理机调度性能的好坏,因而操作系统的调度算法是非常重要的。通过研究基本的操作系统作业（进程）调度算法,详尽分析和对比这些调度算法的优势和劣势。最后对新兴的实时系统研究现状进行介绍和展望,为以后实时系统调度算法研究提供了有效的参考价值。     

实时系统中的调度算法分类研究

本文叙述了实时系统中调度算法的分类、各类算法的研究成果和近期研究状况。     

单机实时系统中的一种混合型实时容错调度算法

目前研究单机实时系统的调度算法文章大多只能调度单一类型的任务。本文在PKSA算法的基础上，建立了一种混合型实时容错模型，提出一种调度算法不仅可以调度有容错需求的周期任务，同时也能够调度无容错需求的周期任务和非周期非实时任务，实现了调度混合型任务的目的。     

面向分布式实时系统的安全驱动调度算法研究

针对现有实时调度算法无法适应动态安全需求的问题,构建了一种安全驱动调度模型,该模型从系统安全级别、系统安全服务和任务安全策略三个方面描述了实时系统的动态安全需求,并设计了一种基于安全驱动的实时任务调度器框架。以该模型和框架为基础,提出了一种安全驱动调度算法(Security Driven Scheduling Algorithm,SDSA)。从全局角度对新到达任务进行可调度性检查,并将可调度任务分配到合适的处理机上运行。按照系统安全级别来动态调整已分配到各处理机上实时任务的安全策略,使其达到安全性和可调度性的最优平衡。采用优先级抢占式策略对各实时任务进行调度。仿真结果表明,SDSA算法与其他同类算法相比,在系统动态安全需求的适应性、关键任务的可调度性以及安全防危能力等方面具有较好的表现。     

异构分布系统的实时任务轮转式容错调度算法

建立了一个异构分布式系统实时调度模型,对异构分布式系统中的任务及不同处理机资源进行了形式化描述.结合基版本/副版本技术,给出了用于异构分布式系统的实时任务轮转式容错调度算法.实例分析表明,该算法有效提高了异构处理机环境下的资源利用率以及整体计算性能.     

On-line scheduling of scalable real-time tasks on multiprocessor systems

The computation time of scalable tasks depends on the number of processors allocated to them in multiprocessor systems. As more processors are allocated to a scalable task, the overall computation time of the task decreases but the total amount of processors’ time devoted to the execution of the task, called workload, increases due to parallel execution overhead. In this paper, we propose a task scheduling algorithm that utilizes the property of scalable tasks for on-line and real-time scheduling. In the proposed algorithm, the total workload of all scheduled tasks is reduced by managing processors allocated to the tasks as few as possible without missing their deadlines. As a result, the processors in the system have less load to execute the scheduled tasks and can execute more newly arriving tasks before their deadlines. Simulation results show that the proposed algorithm performs significantly better than the conventional algorithm based on a fixed number of processors to execute each task.     

实时系统中弹性调度策略

弹性调度面向负载可变的实时系统,通过动态调整任务属性以满足系统的灵活性要求,是一种高效的任务调度策略。针对弹性调度研究中的成果及问题,概述了弹性调度的研究背景,从任务模型、调度模型以及调度算法三个方面对弹性调度的国内外研究进展进行综述,探讨当前研究中存在的问题,并对弹性调度未来研究工作进行分析和展望。     

实时控制任务的抖动研究

在实时控制系统中,由于并发和资源共享等诸多不确定的因素,导致周期任务每次被调度运行的时间有不确定性的变化,系统会变得不稳定。为了保证实时控制系统有较高的性能和可预测性,减小抖动是至关重要的。通过引入截止期缩减因子,对实时系统中抖动敏感的任务确定一个新的相对截止期,能够更好地挖掘出可用的空闲时间,从而使得系统的稳定性提高。实验结果证明提出的方法比文献中用到的其他方法更加有效和可用。     

Power-aware scheduling algorithms for sporadic tasks in real-time systems

In this paper, we consider the canonical sporadic task model with the system-wide energy management problem. Our solution uses a generalized power model, in which the static power and the dynamic power are considered. We present a static solution to schedule the sporadic task set, assuming worst-case execution time for each sporadic tasks release, and propose a dynamic solution to reclaim the slacks left by the earlier completion of tasks than their worst-case estimations. The experimental results show that the proposed static algorithm can reduce the energy consumption by 20.63%–89.70% over the EDF* algorithm and the dynamic algorithm consumes 2.06%–24.89% less energy than that of the existing DVS algorithm.     

移动分布式实时事务实时原子提交

形式地给出了移动分布式实时事务实时原子提交协议的定义,在此基础上提出了适合于移动分布式实时事务的实时原子提交协议：一阶段实时原子提交协议(1PRACP)。1PRACP通过参与者与协调者的一次消息交换,在一个阶段完成移动分布式实时事务提交活动;结合超时恢复处理协议,1PRACP能避免由于站点故障或网络通信链路故障而导致的阻塞。对1PRACP进行了性能比较和评测,显示了它在各方面的优越性。     

Power-aware fixed priority scheduling for sporadic tasks in hard real-time systems

In this paper, we consider the generalized power model in which the focus is the dynamic power and the static power, and we study the problem of the canonical sporadic task scheduling based on the rate-monotonic (RM) scheme. Moreover, we combine with the dynamic voltage scaling (DVS) and dynamic power management (DPM). We present a static low power sporadic tasks scheduling algorithm (SSTLPSA), assuming that each task presents its worst-case work-load to the processor at every instance. In addition, a more energy efficient approach called a dynamic low power sporadic tasks scheduling algorithm (DSTLPSA) is proposed, based on reclaiming the dynamic slack and adjusting the speed of other tasks on-the-fly in order to reduce energy consumption while still meeting the deadlines. The experimental results show that the SSTLPSA algorithm consumes 26.55–38.67% less energy than that of the RM algorithm and the DSTLPSA algorithm reduces the energy consumption up to 18.38–30.51% over the existing DVS algorithm.     

Fault-tolerant scheduling for real-time embedded control systems

With the increasing complexity of industrial application, an embedded control system (ECS) requires processing a number of hard real-time tasks and needs fault-tolerance to assure high reliability. Considering the characteristics of real-time tasks in ECS, an integrated algorithm is proposed to schedule real-time tasks and to guarantee that all real-time tasks are completed before their deadlines even in the presence of faults. Based on the nonpreemptive critical-section protocol (NCSP), this paper analyzes the blocking time introduced by resource conflicts of relevancy tasks in fault-tolerant multiprocessor systems. An extended schedulability condition is presented to check the assignment feasibility of a given task to a processor. A primary/backup approach and on-line replacement of failed processors are used to tolerate processor failures. The analysis reveals that the integrated algorithm bounds the blocking time, requires limited overhead on the number of processors, and still assures good processor utilization. This is also demonstrated by simulation results. Both analysis and simulation show the effectiveness of the proposed algorithm in ECS.     

End-to-end deadline control for aperiodic tasks in distributed real-time systems

A category of Distributed Real-Time Systems (DRTS) that has multiprocessor pipeline architecture is increasingly used. The key challenge of such systems is to guarantee the end-to-end deadlines of aperiodic tasks. This paper proposes an end-to-end deadline control model, called Linear Quadratic Stochastic Optimal Control Model (LQ-SOCM), which features a distributed feedback control that dynamically enforces the desired performance. The control system considers the aperiodic task arrivals and execution times’ variation as the two external factors of the system unpredictability. LQ-SOCM uses discrete time state space equation to describe the real-time computing system. Then, in the actuator design, a continuous manner is adopted to deal with discrete QoS (Quality of Service) adaptation. Finally, experiments demonstrate that the system is globally stable and can statistically provide the end-to-end deadline guarantee for aperiodic tasks. At the same time, LQ-SOCM is capable of effectively improving the system throughput.