Minimizing energy consumption of multiple-processors-core systems with simultaneous task allocation, scheduling and voltage assignment

In this paper, we tackle the problem of minimizing the energy consumption of multiple-processors-core system using multiple variable supply voltages. Given a set of tasks and their dependency, and a set of target processors, we propose an optimal formulation of carrying out task allocation on processors, task scheduling and voltage assignment for each task concurrently using mixed integer nonlinear programming (MINLP). An iterative divide-and conquer approach is introduced to reduce the run time for large set of tasks. Experimental results show that hy using the proposed approach, significant energy reductions are achieved over the existing methods.