Feedback EDF Scheduling of Real-Time Tasks Exploiting Dynamic Voltage Scaling

Abstract

Abstract.Many embedded systems are constrained by limits on power consumption, which are reflected in the design and implementation for conserving their energy utilization. Dynamic voltage scaling (DVS) has become a promising method for embedded systems to exploit multiple voltage and frequency levels and to prolong their battery life. However, pure DVS techniques do not perform well for systems with dynamic workloads where the job execution times vary significantly. In this paper, we present a novel approach combining feedback control with DVS schemes targeting hard real-time systems with dynamic workloads. Our method relies strictly on operating system support by integrating a DVS scheduler and a feedback controller within the earliest-deadline-first (EDF) scheduling algorithm. Each task is divided into two portions. The objective within the first portion is to exploit frequency scaling for the average execution time. Static and dynamic slack is accumulated for each task with slackpassing and preemption handling schemes. The objective within the second portion is to meet the hard real-time deadline requirements up to the worst-case execution time following a lastchance approach. Feedback control techniques make the system capable of selecting the right frequency and voltage settings for the first portion, as well as guaranteeing hard real-time requirements for the overall task. A feedback control model is given to describe our feedback DVS scheduler, which is used to analyze the system's stability. Simulation experiments demonstrate the ability of our algorithm to save up to 29% more energy than previous work for task sets with different dynamic workload characteristics.

Local Adjustment and Global Adaptation of Control Periods for QoC Management of Control Systems

“…Unlike many existing task scheduling and QoC management methods which aim to maintain the processor utilization at a desired value [5], [19], it allows a much lower utilization when one or more control loops have good QoC. This has significant implications to embedded systems with limited power resources [32].…”

A simple power-aware scheduling for multicore systems when running real-time applications

“…Most of them focus on periodic task systems [20,31], but also some proposals have been published dealing with aperiodic tasks [23] and sporadic tasks [22]. Buttazzo et al [20] present an algorithm for energy management based on DVS that integrates the elastic scheduler for discrete voltage mode processors.…”

Reducing jitter in embedded systems employing a time-triggered software architecture and dynamic voltage scaling

“…(Zhu and Mueller, 2004;Zhuo and Chakrabarti, 2005;Cai et al, 2006)) compute voltage settings during runtime to take advantage o f early completion o f tasks (i.e. slack time) in order to increase energy savings.…”

“…Zhu and M ueller (2004) Although general offline and online DVS techniques, e.g. (Gruian, 2001;Shin et a l, 2001;Swaminathan and Chakrabarty, 2001;Kim et al, 2002;Zhu and Mueller, 2004;Wei et al, 2006), are effective in reducing energy consumption, they are not necessarily efficient when it comes to extending battery life because they do not consider non-linear battery characteristics (Cai et al, 2006). Zhuo and Chakrabarti (2005) therefore proposed a battery-aware dynamic task scheduling algorithm, "dynamic average rate EDF" (darEDF), that combines the concepts o f a slack forwarding algorithm (Shin and Choi, 1999) -which tries to make the power profile decrease with time, to improve battery efficiency -and a slack look-ahead algorithm (Shin et a l , 2000), which tries to increase system idle time, to increase opportunities for battery recovery.…”