MPEFT: A novel task scheduling method for workflows

Optimizing the scheduling algorithm is a key problem to improving service efficiency of urban heterogeneous computing platforms. In this paper, we propose novel list-based called Modified Predict Earliest Finish Time (MPEFT) for systems with aim minimize total execution time. The consists two stages: task prioritization and processor selection. phase, priority tasks calculated by time cost all paths from exit task. Compared prior works, more accurate priorities are obtained considering not only critical path but also non-critical ones. selection allocated according whether resources sufficient its successive tasks. experiments on randomly generated workflows practical applications show that MPEFT outperforms other existing list algorithms.