Energy Efficient Scheduling of Independent Tasks on Multicore Processors with Software Controlled Dynamic Voltage Scaling

摘要

Energy-efficient task scheduling for multicore processors is a vital design issue in parallel and distributed computing. Nowadays multicore processors are designed by using software controlled Dynamic Voltage Scaling (DVS) to reduce the energy consumption of cores. We propose a modified EETSA algorithm to schedule the independent tasks on multicore processors that have to be completed within a given common deadline. The proposed algorithm works in two phases. In the first phase, tasks are scheduled on different cores of multicore processors and in the second phase, energy consumption is optimized. The complexity of proposed algorithm is same as the EETSA. We perform the experiments on a system with a single multi-core processor with software controlled DVS having a limited set of core speeds. The results of proposed algorithm are compared with EETSA and OPT algorithms. The results show that proposed algorithm outperforms the compared algorithms regarding scheduling time and energy consumption.