Try-once-discard scheduling for stochastic networked control systems

In practical networked control systems (NCS), such as smart grids, cooperative robotics, and sensor networks, often multiple control applications share a communication infrastructure, requiring a smart and efficient scheduling mechanism to coordinate the access to the capacity-limited communication medium. In this article we consider the problem of event-based scheduling design for NCSs consisting of multiple control loops over a shared communication medium. We extend the notion of Try-Once-Discard (TOD), which is one of the basic deterministic event-based scheduling protocols for resource constrained NCSs, to the case of multiple stochastic control systems coupled via a shared communication medium subject to capacity limitation and stochastic packet delivery failure. Showing that the overall network-induced error is a homogeneous Markov chain in our stochastic set-up, we first study stability properties of such networked systems under the TOD scheduling scheme employing the concepts of stochastic stability. Then, we derive sufficient stability conditions under the TOD rule assuming that the communication channel is not ideal, i.e. a scheduled data packet for transmission might be lost in the communication channel with a non-zero probability. Furthermore, we derive analytic performance bounds by finding uniform upper-bounds for an average quadratic cost function. The numerical simulations are performed for variety of system parameters and NCS set-ups to strengthen our stability claim as well as illustrating performance bounds. Additionally, we show that the TOD scheduling rule outperforms the conventional time-triggered, and uniform and non-uniform random channel access arbitration mechanisms, in terms of efficient coordination of channel access in stochastic NCSs.     

基于机器学习的异构感知多核调度方法

异构多核处理器已成为现代嵌入式系统的主流解决方案，而好的在线映射或调度方法对其充分发挥高性能和低功耗的优势起着至关重要的作用。针对异构多核处理系统上的应用程序动态映射和调度问题，提出一种基于机器学习、能快速准确评估程序性能和程序行为阶段变化的检测技术来有效确定重映射时机从而最大化系统性能的映射和调度解决方案。该方案一方面通过合理选择处理核和程序运行时的静态和动态特征来有效感知异构处理所带来的计算能力和工作负载运行行为的差异，从而能够构建更加准确的预测模型；另一方面通过引入阶段检测来尽可能减少在线映射计算的次数，从而能够提供更加高效的调度方案。最后，在SPLASH-2数据集上验证了所提出调度方案的有效性。实验结果表明，与Linux默认的完全公平调度（CFS）方法相比，所提出的方法在系统计算性能方面提高了52%，在CPU资源利用率上提高了9.4%。这表明所提方法在系统计算性能和CPU资源利用率方面具备优良的性能，可以有效提升异构多核系统的应用动态映射和调度效果。     

基于机器学习的异构感知多核调度方法

异构多核处理器已成为现代嵌入式系统的主流解决方案,而好的在线映射或调度方法对其充分发挥高性能和低功耗的优势起着至关重要的作用。针对异构多核处理系统上的应用程序动态映射和调度问题,提出一种基于机器学习、能快速准确评估程序性能和程序行为阶段变化的检测技术来有效确定重映射时机从而最大化系统性能的映射和调度解决方案。该方案一方面通过合理选择处理核和程序运行时的静态和动态特征来有效感知异构处理所带来的计算能力和工作负载运行行为的差异,从而能够构建更加准确的预测模型;另一方面通过引入阶段检测来尽可能减少在线映射计算的次数,从而能够提供更加高效的调度方案。最后,在SPLASH-2数据集上验证了所提出调度方案的有效性。实验结果表明,与Linux默认的完全公平调度（CFS）方法相比,所提出的方法在系统计算性能方面提高了52%,在CPU资源利用率上提高了9.4%。这表明所提方法在系统计算性能和CPU资源利用率方面具备优良的性能,可以有效提升异构多核系统的应用动态映射和调度效果。     

HSCS: a hybrid shared cache scheduling scheme for multiprogrammed workloads

The traditional dynamic random-access memory (DRAM) storage medium can be integrated on chips via modern emerging 3D-stacking technology to architect a DRAM shared cache in multicore systems. Compared with static random-access memory (SRAM), DRAM is larger but slower. In the existing research, a lot of work has been devoted to improving the workload performance using SRAM and stacked DRAM together in shared cache systems, ranging from SRAM structure improvement to optimizing cache tags and data access. However, little attention has been paid to designing a shared cache scheduling scheme for multiprogrammed workloads with different memory footprints in multicore systems. Motivated by this, we propose a hybrid shared cache scheduling scheme that allows a multicore system to utilize SRAM and 3D-stacked DRAM efficiently, thus achieving better workload performance. This scheduling scheme employs (1) a cache monitor, which is used to collect cache statistics; (2) a cache evaluator, which is used to evaluate the cache information during the process of programs being executed; and (3) a cache switcher, which is used to self-adaptively choose SRAM or DRAM shared cache modules. A cache data migration policy is naturally developed to guarantee that the scheduling scheme works correctly. Extensive experiments are conducted to evaluate the workload performance of our proposed scheme. The experimental results showed that our method can improve the multiprogrammed workload performance by up to 25% compared with state-of-the-art methods (including conventional and DRAM cache systems).     

Feedback scheduling of priority-driven control networks

With traditional open-loop scheduling of network resources, the quality-of-control (QoC) of networked control systems (NCSs) may degrade significantly in the presence of limited bandwidth and variable workload. The goal of this work is to maximize the overall QoC of NCSs through dynamically allocating available network bandwidth. Based on codesign of control and scheduling, an integrated feedback scheduler is developed to enable flexible QoC management in dynamic environments. It encompasses a cascaded feedback scheduling module for sampling period adjustment and a direct feedback scheduling module for priority modification. The inherent characteristics of priority-driven control networks make it feasible to implement the proposed feedback scheduler in real-world systems. Extensive simulations show that the proposed approach leads to significant QoC improvement over the traditional open-loop scheduling scheme under both underloaded and overloaded network conditions.     

Structural Properties,LQG Control and Scheduling of a Networked Control System with Bandwidth Limitations and Transmission Delays

The problem of simultaneous LQG control and scheduling of a Networked Control System (NCS) with constant network induced delays at input and output and bandwidth limitations is investigated. Delays are considered at plant as well as controller side. Sufficient conditions for controllability, stabilizability, reconstructibility and detectability of the underlying networked control system are drawn. The proposed conditions extend previous works on structural properties of NCS by capturing both plant and controller side delays together with bandwidth limitations. A framework for computing the optimal LQG controller for the NCS with a fixed scheduling is provided. The proposed modeling approach facilitates use of LQG as well as other control methods for NCSs with delays and bandwidth limitations. In order to optimize performance, a semi-online scheduling procedure is proposed based on an offline look up table. The look up table assigns an optimal schedule with associated optimal LQG controller to initial conditions. The proposed scheme improves previous results by online deployment of schedule and LQG control with stability guarantees and very low computational overhead. A simulation example with communication delays, packet losses and bandwidth limitations in both sensor and actuator sides is included. Static optimal periodic communication sequence, Optimal Pointer Placement (OPP) approach proposed in previous works, a random access scheduling method representing contention based access policies and the proposed method are simulated and compared.      

Dual scheduling and quantised control for networked control systems with communication constraints

A novel integrated design scheme of average dwell time scheduling strategy, dynamic bandwidth allocation policy and quantised control for a collection of networked control systems (NCSs) with time delay and communication constraints is proposed in this paper. A scheduling policy is presented to accommodate the limitation of communication capacity which depends on the convergence rate of closed-loop system and divergence rate of open-loop plant. Linear programming technique is adopted to dynamically allocate bit rate for each node and the strategy is used to make trade-offs between the network utilisation and the control performance which provides an effective way of optimising the quality of control (QoC) and the quality of service (QoS) for NCSs. Mid-tread uniform quantisers update the quantisation rules according to the assignment of the bit rate and convert the quantised state into a kind of input saturation with bounded disturbances. Taking into account the effect of dual scheduling strategy and quantisation, the NCSs are modelled as discrete-time switched systems with bounded disturbances. Furthermore, a scheduling and quantised feedback control co-design procedure is proposed for the simultaneous stabilisation of the collection of networked subsystems. Finally, a simulation example is given to illustrate the effectiveness of the proposed method.     

网络控制系统中的调度问题

网络控制系统（NCS）的控制性能受限于共享的计算资源和网络资源,如何对NCS中的周期和非周期的复杂信息流进行有效的调度成为影响系统控制性能的关键因素。从实时控制角度,分析了NCS中的时态特性和NCS的实时调度问题。同时从实时CPU调度策略、NCS网络调度方法、调度诱导问题及其补偿方法、可调度性分析和调度优化、控制与调度协同设计等方面综述了现有的NCS中的调度理论和方法,并讨论了今后进一步的研究方向。     

An energy-efficient real-time scheduling scheme on dual-channel networks

The recent evolution of wireless sensor networks have yielded a demand to improve energy-efficient scheduling algorithms and energy-efficient medium access protocols. This paper proposes an energy-efficient real-time scheduling scheme that reduces power consumption and network errors on dual channel networks. The proposed scheme is based on a dynamic modulation scaling scheme which can scale the number of bits per symbol and a switching scheme which can swap the polling schedule between channels. Built on top of EDF scheduling policy, the proposed scheme enhances the power performance without violating the constraints of real-time streams. The simulation results show that the proposed scheme enhances fault-tolerance and reduces power consumption.     

Energy-efficient deadline scheduling for heterogeneous systems

Energy efficiency is a major concern in modern high performance computing (HPC) systems and a power-aware scheduling approach is a promising way to achieve that. While there are a number of studies in power-aware scheduling by means of dynamic power management (DPM) and/or dynamic voltage and frequency scaling (DVFS) techniques, most of them only consider scheduling at a steady state. However, HPC applications like scientific visualization often need deadline constraints to guarantee timely completion. In this paper we present power-aware scheduling algorithms with deadline constraints for heterogeneous systems. We formulate the problem by extending the traditional multiprocessor scheduling and design approximation algorithms with analysis on the worst-case performance. We also present a pricing scheme for tasks in the way that the price of a task varies as its energy usage as well as largely depending on the tightness of its deadline. Last we extend the proposed algorithm to the control dependence graph and the online case which is more realistic. Through the extensive experiments, we demonstrate that the proposed algorithm achieves near-optimal energy efficiency, on average 16.4% better for synthetic workload and 12.9% better for realistic workload than the EDD (Earliest Due Date)-based algorithm; The extended online algorithm also outperforms the EDF (Earliest Deadline First)-based algorithm with an average up to 26% of energy saving and 22% of deadline satisfaction. It is experimentally shown as well that the pricing scheme provides a flexible trade-off between deadline tightness and price.     

Safe self-scheduling: A parallel loop scheduling scheme for shared-memory multiprocessors

In this papaer was present Safe Self-Scheduling (SSS), a new scheduling scheme that schedules parallel loops with variable length iteration execution times not known at compile time. The scheme assumes a shared memory space. SSS combines static scheduling with dynamic scheduling and draws favorable advantages from each. First, it reduces the dynamic scheduling overhead by statically scheduling a major portion of loop iterations. Second, the workload is balanced with a simple and efficient self-scheduling scheme by applying a new measure, thesmallest critical chore size. Experimental results comparing SSS with other scheduling schemes indicate that SSS surpasses other scheduling schemes. In the experiment on Gauss-Jordan, an application that is suitable for static scheduling schemes, SSS is the only self-scheduling scheme that outperforms the static scheduling scheme. This indicates that SSS achieves a balanced workload with a very small amount of overhead. This research has been supported in part by the National Science Foundation under Contract No. CCR-9210568.     

Security‐aware scheduling model for computational grid

Grid applications with stringent security requirements introduce challenging concerns because the schedule devised by nonsecurity‐aware scheduling algorithms may suffer in scheduling security constraints tasks. To make security‐aware scheduling, estimation and quantification of security overhead is necessary. The proposed model quantifies security, in the form of security levels, on the basis of the negotiated cipher suite between task and the grid‐node and incorporates it into existing heuristics MinMin and MaxMin to make it security‐aware MinMin(SA) and MaxMin(SA). It also proposes SPMaxMin (Security Prioritized MinMin) and its comparison with three heuristics MinMin(SA), MaxMin(SA), and SPMinMin on heterogeneous grid/task environment. Extensive computer simulation results reveal that the performance of the various heuristics varies with the variation in computational and security heterogeneity. Its analysis over nine heterogeneous grid/task workload situations indicates that an algorithm that performs better for one workload degrades in another. It is conspicuous that for a particular workload one algorithm gives better makespan while another gives better response time. Finally, a security‐aware scheduling model is proposed, which adapts itself to the dynamic nature of the grid and picks the best suited algorithm among the four analyzed heuristics on the basis of job characteristics, grid characteristics, and desired performance metric. Copyright © 2011 John Wiley & Sons, Ltd.     

基于动态调度的网络控制系统设计方法

针对网络控制系统中存在多回路共享网络资源的问题,本文提出一种基于反馈和预测机理的动态调度策略,根据当前的网络状态在线调整控制系统的采样周期,以适应当前网络中信息流的变化,达到改善网络运行性能的目的．采用动态调度策略的网络控制系统为不确定变采样周期系统,经过等效变换,将系统等效为一类具有参数不确定性的离散系统．基于建立的模型,给出系统的鲁棒控制器参数化设计方法,并通过仿真实例验证了所提方法的有效性．     

Guaranteed cost control for networked control systems

The guaranteed cost control problem for networked control systems (NCSs) is addressed under conmmnication constraints and varying sampling rate. First of all, a simple inFormation-scheduling scheme is presented to describe the scheduling approach of system signals in NCSs. Then, based on such a scheme and given sampling method, the design procedure in dynarmic output feedback manner is also derived which renders the closed loop system to be asymptotically stable and guarantees an upper bound of the LQ pefformance cost function.     

Information delivery through broadcasting in satellite communication networks

Satellite broadcast is an important candidate for large-scale multimedia information distribution due to the inherent wide-range multicasting capability of satellites and the asymmetry of satellite communications (high bandwidth downlink, limited bandwidth uplink) that matches nicely the information flow asymmetry in multimedia applications. We consider a data broadcasting model that is encountered in most asymmetric satellite communication environments. The problem of scheduling the data broadcast such that average response time experienced by the users is low is considered. In a push-based system, where the users cannot place requests directly to the server and the broadcast schedule should be determined based solely on the access probabilities, we formulate a deterministic dynamic optimization problem, the solution of which provides the optimal broadcast schedule. Properties of the optimal solution are obtained and then we propose a suboptimal dynamic policy which achieves average response time close to the lower bound. In a pull-based system where the users may place requests about information items directly to the server, the scheduling can be based on the number of pending requests for each item. Suboptimal policies with good performance are obtained in this case as well. If a user has local memory, it can alleviate its access latency by selectively prefetching the items from the broadcast and storing them in the memory. A good memory management strategy can substantially reduce the user's access latency. An optimal memory management policy is identified, that minimizes the expected aggregate latency. Memory update strategies with limited look-ahead are presented as implementable approximations of the optimal policy as well. We also consider the problem of joint broadcast scheduling and user's cache management and propos a joint optimization scheme which can achieve the performance up to 40% better than the existing non-joint approach.     

数据流程序动态调度与优化

为了解决数据流编程模型的可用性问题,使其能在兼顾程序并行性的前提下适用于动态数据交互速率的流应用,设计了一种动态调度与静态优化相结合的数据流编译系统。编译器以COStream语言编写的源程序为输入,通过对源程序进行分析,以动态速率的数据通信边作为边界划分程序到粗粒度的子图,在子图内部应用静态优化。根据子图的每个计算单元的工作量估计计算资源的使用状况,实现子图内计算单元到处理器核的映射,经过阶段划分分配子图内计算单元到相应流水阶段。在运行时,每个子图在各个处理器核上均启动一个线程,通过对线程间通信的优化,避免了运行时多个线程对同一段内存同时读写产生的同步开销,减少了线程的上下文切换次数。使用信号量控制子图内线程间的同步,基于各子图计算单元运行时数据交互速率并结合当前线程的状态,动态调度各个子图的执行,构建动态的软件流水线,生成相应多线程目标代码。实验以通用X86-64多核处理器作为实验平台,测试和分析数据流编译的性能。实验结果表明,编译系统可以实现动态数据交互速率的数据流应用,扩大了编译系统可用性并且具有一定加速效果。     

Toward multi-programmed workloads with different memory footprints: a self-adaptive last level cache scheduling scheme

With the emerging of 3D-stacking technology, the dynamic random-access memory (DRAM) can be stacked on chips to architect the DRAM last level cache (LLC). Compared with static randomaccess memory (SRAM), DRAM is larger but slower. In the existing research papers, a lot of work has been devoted to improving the workload performance using SRAM and stacked DRAM together, ranging from SRAM structure improvement, to optimizing cache tag and data access. Instead, little attention has been paid to designing an LLC scheduling scheme for multi-programmed workloads with different memory footprints. Motivated by this, we propose a self-adaptive LLC scheduling scheme, which allows us to utilize SRAM and 3D-stacked DRAM efficiently, achieving better workload performance. This scheduling scheme employs (1) an evaluation unit, which is used to probe and evaluate the cache information during the process of programs being executed; and (2) an implementation unit, which is used to self-adaptively choose SRAM or DRAM. To make the scheduling scheme work correctly, we develop a data migration policy. We conduct extensive experiments to evaluate the performance of our proposed scheme. Experimental results show that our method can improve the multi-programmed workload performance by up to 30% compared with the state-of-the-art methods.     

Enhancing WLAN/UMTS dual-mode services using a novel distributed multi-agent scheduling scheme

WLAN services are inexpensive and have a high bandwidth, while UMTS services provide wider coverage area and high mobility. Based on intelligent deduction, this study presents a novel service scheduling scheme for WLAN/UMTS dual-mode networks. The proposed system, Distributed Multi-Agent System (DMAS), consists of a set of problem-solving agents that autonomously process their own tasks and interoperate with one another by a shared database to reach a suitable schedule for dual-mode network services. A two-level control mechanism comprising local-control and meta-control is presented to achieve a high degree of goodness in service scheduling. Simulation results indicate that the Quality of Service (QoS) of the proposed discipline in terms of average delay and jitter is better than that of the pure UMTS network by 25% and 10%, respectively. The scheduling discipline can improve the service quality in multiple-mode networks.     

基于TDMA方式WMN中一种链路调度机制研究

基于TDMA方式的无线网状网中,链路调度对网络性能起着重要作用.针对固定顺序的待调度链路集,提出求解最优调度周期的启发式算法;基于链路顺序对算法性能的影响,从全局优化的角度对全网链路进行排序,提出基于遗传算法的最优链路调度机制.仿真结果表明,该算法能快速收敛于全网链路的最小调度周期,具有比现有算法更高的传输效率和更低的实施复杂度.     

基于强化学习方法的访存调度算法

在现代处理器中,存储控制器是处理器芯片对片外存储器进行访问的管理者和执行者,其中对访存过程的调度算法会对实际访存性能产生十分重要的影响。针对已有调度算法在不同负载特征下自适应性不足的问题,提出了一种基于强化学习方法的ALHS算法,通过对访存调度中页命中优先时的连续页命中上限次数进行自适应调整,习得最优策略。多种不同典型访存模式的模拟结果显示,相比传统的FR-FCFS,ALHS算法运行速度平均提升了10.98%,并且可以获得近似于最优策略的性能提升,表明该算法能够自主探索环境并自我优化。