Skip to Main Content
Efficient task scheduling is critical for achieving high performance in heterogeneous computing systems. Many previous relevant works for Out-Tree task graphs focused on homogeneous environments, while neglecting the heterogeneity of processors and the economization on processors, which resulted in low practical efficiency. This paper presents a heuristic greedy algorithm based on list and task duplication for scheduling Out-Tree task graphs in heterogeneous computing systems, which tries to find the best point between balancing loads and shortening the schedule length and improves the schedule performance without increasing the time complexity of the algorithm. The comparative experimental results demonstrate that the proposed algorithm could achieve shorter schedule length while using less number of processors.