Optimal resource allocation for stochastic systems performance optimisation of control tasks undergoing stochastic execution times (original) (raw)
Related papers
Optimal CPU Allocation to a Set of Control Tasks with Soft Real–Time Execution Constraints ∗
2013
We consider a set of control tasks sharing a CPU and having stochastic execution requirements. Each task is associated with a deadline: when this constraint is violated the particular execution is dropped. Different choices of the scheduling parameters correspond to a different probability of deadline violation, which can be translated into a different level for the Quality of Control experienced by the feedback loop. For a particular choice of the metric quantifying the global QoC, we show how to find the optimal choice of the scheduling parameters.
Journal of Parallel and Distributed Computing, 2016
Heterogeneous parallel and distributed computing systems frequently must operate in environments where there is uncertainty in system parameters. Robustness can be defined as the degree to which a system can function correctly in the presence of parameter values different from those assumed. In such an environment, the execution time of any given task may fluctuate substantially due to factors such as the content of data to be processed. Determining a resource allocation that is robust against this uncertainty is an important area of research. In this study, we define a stochastic robustness measure to facilitate resource allocation decisions in a dynamic environment where tasks are subject to individual hard deadlines and each task requires some input data to start execution. In this environment, the tasks that cannot meet their deadlines are dropped (i.e., discarded). We define methods to determine the stochastic completion times of tasks in the presence of the task dropping. The stochastic task completion time is used in the definition of the stochastic robustness measure. Based on this stochastic robustness measure, we design novel resource allocation techniques that work in immediate and batch modes, with the goal of maximizing the number of tasks that meet their individual deadlines. We compare the performance of our technique against several well-known approaches taken from the literature and adapted to our environment. Simulation results of this study demonstrate the suitability of our new technique in a dynamic heterogeneous computing system.
Scheduling of stochastic tasks on two parallel processors
Naval Research Logistics Quarterly, 1979
We consider the problem o f scheduling n tasks on two identical parallel processors. We show both in tlie case when tlie processing times for the f? tasks are independent exponential random varinbles, and when they are independent liyperexponenti~ls which are mixtures of two fixed exponentials, tlia! tile policy of performing tasks with longest expected processing time (LEPT) first minimiles the expected makespiin, and that in the hyperexponentiill case t h e policy o f performing ksks \\ ith shortest expected processing time (SEPT) lirst minimizes the expected l l o w time. The approach is simpler than the dynamic progrttmming approach recently employed by Bruno and Downey.
Batch Mode Stochastic-Based Robust Dynamic Resource Allocation in a Heterogeneous Computing System
Parallel and Distributed Processing Techniques and Applications, 2010
Heterogeneous, parallel and distributed computing systems frequently must operate in environments where uncertainty in system parameters is common. Robustness can be defined as the degree to which a system can function correctly in the presence of parameter values different from those assumed. In such an environment, the amount of processing required to complete any given task may fluctuate substantially due to variations in data size and content. Determining a resource allocation that accounts for this uncertainty is an important area of research. In this study, we define a stochastic robustness measure to facilitate batchmode resource allocation decisions in a dynamic environment where tasks are subject to individual deadlines and design a novel resource allocation technique that attempts to maximize our new stochastic robustness measure. We compare the performance of our technique against some commonly used approaches taken from the literature and adapted to our environment. Our performance results demonstrate the viability of our new technique in a dynamic heterogeneous computing system.
Optimal scheduling of control tasks with state feedback resource allocation
2006 American Control Conference, 2006
In a large category of embedded systems, computing resources are limited. Consequently, they need to be exploited as efficiently as possible. Recently, many research works have demonstrated that considering jointly the problems of control and scheduling leads to a better control performance, given the same computing resources. In this paper, the problem of the optimal integrated control and non-preemptive off-line scheduling of control tasks in the sense of the H 2 performance criterion is addressed. It is shown that this problem can be decomposed into two sub-problems which can be solved separately. The first sub-problem aims at finding the optimal non-preemptive off-line schedule, and is solved using efficient Branch and Bound algorithms. The second sub-problem uses the lifting technique to determine the optimal control gains, based on the solution of the first sub-problem. Finally, in order to improve the control performance by dynamically allocating the computational resources, an efficient on-line scheduling heuristic is proposed.
A General Framework for Scheduling in a Stochastic Environment
2007
There are many systems and techniques that address stochastic scheduling problems, based on distinct and sometimes opposite approaches, especially in terms of how scheduling and schedule execution are combined, and if and when knowledge about the uncertainties are taken into account. In many real-life problems, it appears that all these approaches are needed and should be combined, which to our knowledge has never been done. Hence it it first desirable to define a thorough classification of the techniques and systems, exhibiting relevant features: in this paper, we propose a tree-dimension typology that distinguishes between proactive, progressive, and revision techniques. Then a theoretical representation model integrating those three distinct approaches is defined. This model serves as a general template within which parameters can be tuned to implement a system that will fit specific application needs: we briefly introduce in this paper our first experimental prototypes which validate our model.
Optimal computational resource allocation for control task under fixed priority scheduling
IFAC Proceedings Volumes (IFAC-PapersOnline), 2011
In this paper a new real-time control system co-design is presented. Given several plants controlled by Linear Quadratic Regulator algorithms running over the same real-time platform, an optimal selection method of the sampling time of each regulator is proposed. The method finds the optimal solution that minimizes an appropriate overall cost function taking into account performance of each subsystem within a constraint on the computational resource. To deal with this problem a new statespace approach for modeling systems with any value of computational delay is proposed. The structure of the optimization problem is then exploited in a way that its solution can be found by solving a minimal set of nonlinear equations. A simple example with three subsystems is presented to highlight the main features of the proposed method.
Energy and Deadline Constrained Robust Stochastic Static Resource Allocation
Lecture Notes in Computer Science, 2014
In this paper, we study the problem of energy and deadline constrained static resource allocation where a collection of independent tasks ("bag-of-tasks") is assigned to a heterogeneous computing system. Computing systems often operate in environments where task execution times vary (e.g., due to data dependent execution times), therefore we model the execution time of tasks stochastically. This research focuses on the design of energy-constrained resource allocation heuristics that maximize robustness against the uncertainties in task execution times. We design and evaluate a new resource allocation heuristic based on Tabu Search that employs dynamic voltage and frequency scaling (DVFS) and exploits heterogeneity by incorporating novel local search techniques.
An efficient resource allocation approach in real-time stochastic environment
2006
Abstract We are interested in contributing to solving effectively a particular type of real-time stochastic resource allocation problem. Firstly, one distinction is that certain tasks may create other tasks. Then, positive and negative interactions among the resources are considered, in achieving the tasks, in order to obtain and maintain an efficient coordination. A standard Multiagent Markov Decision Process (MMDP) approach is too prohibitive to solve this type of problem in real-time.
Feedback scheduling of control tasks
Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187), 2000
The paper presents a feedback scheduling mechanism in the context of co-design of the scheduler and the control tasks. We are particularly interested in controllers where the execution time may change abruptly between different modes, such as in hybrid controllers. The proposed solution attempts to keep the CPU utilization at a high level, avoid overload, and distribute the computing resources evenly among the tasks. The feedback scheduler is implemented as a periodic or sporadic task that assigns sampling periods to the controllers based on execution-time measurements. The controllers may also communicate feedforward mode-change information to the scheduler. As an example, we consider hybrid control of a set of double-tank processes. The system is evaluated, from both scheduling and control performance perspectives, by co-simulation of controllers, scheduler, and tanks.