Predictable execution adaptivity through embedding dynamic reconfigurability into static MPSoC schedules

Advances in semiconductor technologies have placed MPSoCs center stage as a standard architecture for embedded applications of ever increasing complexity. Because of real-time constraints, applications are usually statically parallelized and scheduled onto the target MPSoC so as to obtain predictable worst-case performance. However, both technology scaling trends and resource competition among applications have led to variations in the availability of resources during execution, thus questioning the dynamic viability of the initial static schedules. To eliminate this problem, in this paper we propose to statically generate a compact schedule with predictable response to various resource availability constraints. Such schedules are generated by adhering to a novel band structure, capable of spawning dynamically a regular reassignment upon resource variations. Through incorporating several soft constraints into the original scheduling heuristic, the proposed technique can furthermore exploit the inherent timing slack between dependent tasks, thus retaining the spatial and temporal locality of the original schedule. The efficacy of the proposed technique is confirmed by incorporating it into a widely adopted list scheduling heuristic, and experimentally verifying it in the context of single processor deallocations.