Heterogeneous Processor Scheduling Using Adaptive Particle Swarm Optimization For DVFS Enabled Embedded Systems

Abstract

In this work, we have focused on minimizing the energy consumption for Embedded systems and improving the resource utilization. Currently, DVFS based schemes are widely adopted in various applications but these techniques are based on scaling the voltage and frequencies dynamically, but when the number of tasks are increased then these schemes fail to handle the scenario and degrades the system performance in terms of resource and energy consumption. In order to overcome these issues, we present a novel hybrid approach by combining DVFS and Particle swarm optimization-based scheme for processor scheduling. In this approach, the task finish time and energy consumption are considered as the objective function. The proposed approach is implemented using MATLAB simulation tool and obtained performance is compared with state-of-art techniques. The comparative study shows, the proposed approach achieves better performance in terms of minimizing the energy consumption.