Thermally Optimal Stop-Go Scheduling of Task Graphs with Real-Time Constraints

Pratyush Kumar, Lothar Thiele.
ASP-DAC, Jan. 25-28 2011, Yokohama, Japan.
Dynamic thermal management (DTM) techniques to manage the load on a system to avoid thermal hazards are soon becoming mainstream in today’s systems. Several research works have studied the use of dynamic voltage scaling (DVS) to serve a set of independent real-time tasks with temperature/performance constraints. In this paper, we study another important class of DTM techniques: stop-go scheduling, to minimize peak temperature when scheduling an application modelled as a task-graph within a given makespan constraint. For a given static-ordering of execution of tasks, we derive the optimal schedule referred to as a JUST policy. We also prove that for periodic task-graphs the optimal temperature is independent of the chosen static-ordering. Simulation experiments validate the theoretical results.