Abstract:

The workflow scheduling problem in heterogeneous distributed systems is hard to solve due to both the intermediate data transfer time and the computation time for each task being considered.The paper has a study of the dataaware workflow scheduling algorithm based on Astar,to achieve optimal scheduling which is through the overlapping of task execution and data deployment on computing sites.The simulation results show that, in most cases,the improved algorithm is superior to the existing work in performance and efficiency,and significantly reduces the turnaround time.In addition,we also extend the algorithm to solve the process coscheduling problem.

Key words: workflow scheduling；largescale scientific computing；optimal scheduling；turnaround time