Real-time scheduling algorithms proposed in the
literature are often based on worst-case estimates of task
parameters and the performance of an open-loop scheme can
therefore be poor. To improve on such a situation, one can
instead apply a closed-loop scheme, where feedback is exploited
to dynamically adjust the system parameters at run-time. We
propose an optimal control framework that takes advantage
of feeding back information of finished tasks to solve a realtime
multiprocessor scheduling problem with uncertainty in
task execution times, with the objective of minimizing the
total energy consumption. Specifically, we propose a linear
programming-based algorithm to solve a workload partitioning
problem and adopt McNaughton’s wrap around algorithm
to find the task execution order. Simulation results for a
PowerPC 405LP and an XScale processor illustrate that our
feedback scheduling algorithm can result in an energy saving
of approximately 40% compared to an open-loop method.