LAUSR

Harvesting renewable generation (e.g., solar energy) from the ambient environment to achieve a near perpetual operation for embedded systems is being paid more and more attention by academia and industry. However, an immediate problem along with the utilization of renewable energy is the degraded system throughput caused by the intermittent characteristic of renewable generation. On the other hand, energy-harvesting systems (EHSs) deployed in harsh environment are more vulnerable to transient and permanent faults. This paper aims at scheduling dependent tasks on a multicore platform for throughput maximization under energy and reliability constraints. The target of this paper is to design algorithms that optimize system throughput under the energy, reliability, as well as task precedence constraints. To achieve this goal, we propose a mixed-integer linear programming (MILP) approach for allocating and scheduling precedence constrained tasks on the multicore to maximize the throughput of EHS. However, the MILP may find the optimal solution in an exponential time. To overcome this difficulty, we propose a polynomial-time heuristic algorithm to solve the MILP-based throughput maximization problem. In this heuristic algorithm, the uncertainty in energy sources is considered and the allocation and scheduling of tasks are determined based on system energy state. The extensive simulation experiments are carried out to validate our MILP approach and throughput-aware heuristic algorithm. The simulation results justify that the MILP approach achieves an up to 92.9% improvement of system throughput when compared with a baseline method, and the proposed heuristic improves system throughput by up to 32.1% on average when compared with the four representative existing approaches.