Battery Aware Dynamic Scheduling for Periodic Task Graphs

Battery lifetime is a primary design constraint for mobile embedded systems. It has been shown to depend heavily on the load current profile (i.e. evolution of the current drawn over time). However, up to now, very few low-power scheduling policies take this fact into account. We explore how scheduling guidelines drawn from battery models can help in the extension of battery capacity. We proposed a 'Battery-Aware Scheduling' methodology for periodically arriving task-graphs (Directed Acyclic Graph) with real time deadlines and precedence constraints. The methodology presented divides the problem into two steps. First, a good DVS algorithms dynamically determines the minimum frequency of execution. Then, a greedy algorithm allows a near optimal priority function to choose the task which would maximize slack recovery. Battery simulations carried out on the profile generated by our approach for a large set of task-graphs show that battery life time is extended up to 23.3% compared to existing dynamic scheduling schemes