A Study of Cloud Computing Scheduling Algorithm Based on Task Decomposition

Abstract

The quality of cloud computing scheduling algorithm design can leave great impact on arithmetic speed. The key of restricting scheduling algorithm lies in task decomposition. This thesis conducted a study on task decomposition in cloud computing and applied a heuristic approach to design a task decomposition algorithm based on and-or-tree ideology. Also, it did a simulation test on this algorithm. The results of simulation test showed that: compared with those of traditional task decomposition algorithm, there are significant advantages of average handling capacity, time delay, average execution time, and task loss rate contained in the improved algorithm designed in this thesis.