基于粒子群优化的多核处理器系统节能调度算法  

Energy-Saving Scheduling Algorithm of Multi-Core Processor System Based on Particle Swarm Optimization

作　　者：赵阳[1] 满君丰[1] 龙永新[1] 郭德彪[1] 易旺[1] 

机构地区：[1]湖南工业大学计算机与通信学院,湖南株洲412007

出　　处：《湖南工业大学学报》2012年第6期51-54,82,共5页Journal of Hunan University of Technology

摘　　要：针对嵌入式多核系统中的实时性和能耗问题,提出了基于粒子群优化的多核处理器系统节能调度算法。通过对多核处理器系统任务调度和能量消耗的分析,建立了新的编码策略和相应的目标函数,将任务划分、任务调度及电压选择3个过程整合到粒子群算法的一个单迭代寻优循环中,并兼顾系统的实时性,将问题转化为有约束的粒子群优化问题,利用可行性规则的约束处理技术,形成基于约束粒子群优化的整体节能调度算法。实验结果表明:本文算法能以较快的速度收敛于使系统能耗更少的调度解。In order to solve the problems of real-time and energy consumption for embedded multi-core system, an energy-saving scheduling algorithm based on particle swarm optimization is proposed. According to the analysis of task scheduling and energy consumption for multi-core processor system, a new coding strategy and the corresponding target function are established. Three processes of task partition, task scheduling and voltage selection of energy-efficient scheduling for multi-core processor system are integrated into a simple iterative optimization cycle of the particle swarm algorithm, and compatible with the system real-time, the original issues are transformed into new ones of constrained particle swarm optimization, and a constraint processing technology based on feasibility rules is introduced. Thus a new overall energy-saving scheduling algorithm is gained based on constrained particle swarm optimization. The simulated results show that the proposed algorithm can rapidly converge to the optimal scheduling solution of less energy consumption.

关 键 词：嵌入式多核处理器系统 实时性 能耗 粒子群优化 约束优化 

分 类 号：TP303[自动化与计算机技术—计算机系统结构]