Response Time Analysis of Parallel Tasks under Global Fixed-priority on Heterogeneous Multi-cores

Abstract

Heterogeneous multi-core platforms can provide diverse computing power. By using different cores to perform specific workloads, the advantages of heterogeneous multi-cores can be fully utilized to achieve high performance and low power consumption. In this paper, we study the problem of schedulability analysis of parallel directed acyclic graph (DAG) tasks on heterogeneous platforms. The DAG task studied in this paper has particularities. A vertex of a DAG task is only allowed to execute on a specific set of cores. In this paper, we study each DAG task’s worst-case response time(WCRT) analysis problem and propose an analyzed method for this particular task model. The new method profoundly looks at the higher priority tasks and combines them with the task graph structure information. Searching for the graph structure is time-consuming, so we propose an efficient analysis method that can be performed in polynomial time. We use simulation experiments to verify the accuracy and time consumption of the proposed algorithm. And the results show that our proposed method can significantly improve the existing method.