Energy-Efficient Task Scheduling and Synchronization for Multicore Real-Time Systems

摘要

In the recent years, many energy-efficient realtime task scheduling algorithms have been proposed for multicore platforms. Most of them focus on independent tasks, however, relatively little work has been done in the presence of task synchronization. In this paper, we are interested in scheduling of real-time tasks which may make requests for accessing shared resources at the run time. A similarity-based task-to-core partitioning algorithm is proposed to allocate the tasks which may access a similar set of shared resources to the same core so that a large number of blockings can be avoided. Furthermore, we also propose speed assignment methods to assign proper processor speeds for tasks' executions with full-chip and per-core DVFS techniques so that the overall energy consumption can be reduced. The schedulability analysis of our proposed approach is provided in this paper. The capabilities of our proposed approach were evaluated by a series of experiments for which we have some encouraging results.