A power-aware optimal processor cycle allocation scheme with consideration of fair quality of service (QoS) guarantee in real-time systems is formulated. Although power-aware resource management using dynamic voltage and frequency scaling techniques can be used to resolve the trade-off between QoS maximization and energy consumption minimization, it is also necessary to ensure fair QoS allocation in real-time resource allocation to avoid deterioration of QoS levels. The formulation presented in this paper is a multiple-objective optimization problem in which the objective functions are minimization of total enemy consumption and maximization of the QoS level. A Pareto optimal solution representing the trade-off between the two objective functions is obtained numerically using a weighted-sum method for the case that the resource consumption functions are convex. We show an example in cast that all resource consumption functions are linear
Zhiquan BaiTongtong WangPiming MaYanbo MaKyung Sup Kwak
Ailing ZhongDapeng WuBoran YangRuyan Wang
Maosheng ZhuXi LiHong JiHeli Zhang
Xiaolei ChenAihua ZhangXinzhu Yang