A hybrid framework for application allocation and scheduling in multicore systems with energy harvesting

Abstract

In this paper, we propose a novel hybrid design-time and run-time framework for allocating and scheduling applications in multi-core embedded systems with solar energy harvesting. Due to limited energy availability at run-time, our framework offloads scheduling complexity to design time by creating energy-efficient schedule templates for varying energy budget levels, which are selected at run-time in a manner that is contingent on the available harvested energy and executed with a lightweight slack reclamation scheme that extracts additional energy savings. Our experimental results show that the proposed framework produces energy-efficient and dependency-aware schedules to execute applications under varying and stringent energy constraints, with 23-40% lower miss rates than in prior works on harvesting energy-aware scheduling.