CETAS
Yanshul Sharma, Sanjay Moulik
Abstract
Over the years, the nature of processing platforms is witnessing a significant shift in most of the battery supported real-time systems, which currently underpins a blend of specific multicores to satisfy the needs of present day applications. Devising energy and temperature efficient schedulers has become a critical issue for such kind of systems. Hence, this research presents a heuristic strategy named, CETAS, for energy and temperature efficient scheduling of a set of real-time periodic tasks on a DVFS enabled heterogeneous platform. The presented strategy operates in four stages, namely Deadline Partitioning, Core Clustering, Temperature-Aware Scheduling and Energy-Aware Scheduling. Our experimental analysis shows that CETAS is not only able to improve success ratios for task sets (as high as 3.44%) compared to state-of-the-art but also improve energy savings (as high as 10.71%) as well as reduce temperature (as high as 7.14%) in the system.