Cloud computing task scheduling research based on dynamic programming
Yanbing Liu · Journal of Chongqing University of Posts and Telecommunications · 2012
Task scheduling is the core issue in cloud computing system,the aim of task scheduling is reasonably distributing tasks to achieve the optimal scheduling scheme and complete tasks efficiently.In this paper,we have firstly analyzed existing task scheduling algorithms in cloud computing and pointed out the shortages and disadvantages of the traditional algorithms,then under these conditions we propose a task scheduling algorithm based on dynamic programming model.We think the match between tasks and virtual machines in the datacenter cluster is a problem about combination optimization of multi-stage decision-making,and its objective is the least execution time.In the end,max-min,min-min algorithms and the proposed algorithm are simulated with the aid of the CloudSim toolkit.The result demonstrates that the proposed algorithm has better performance in terms of task completing time and resource load than the classical algorithms.