Efficient energy constrained scheduling approach for dynamic real time system

This paper aims to present a general scheduling algorithm that offers lesser energy consumption for battery powered dynamic real time system modeled with aperiodic tasks by utilizing the concept of dynamic voltage scaling. In this paper we show that greedy based (non lazy approach) are not suitable for limited source of energy constraint. Based on this motivation we proposed window based lazy scheduling algorithm that find the tradeoff between timing constraint as well as available energy. We further present speed stretching algorithm to assign the speed of aperiodic task in such a way that overall energy consumption is reduced and accepts more number of task to improve the quality of service as well as system utilization. The extensive examples and simulation results illustrate that our approach can effectively reduce the energy consumption leading to reduction in battery size.