Cloud service selection optimization method based on parallel discrete particle swarm optimization
Yimin Zhang, Sheng Guojun, Yang Xiaoguang · 2018
With the arrival of the era of the Internet cloud, cloud service composition "explosive" increase brings on a typical NP hard problem of the cloud service selection, the problem can not be solved by the traditional method of integer programming or linear programming. This paper introduces parallel discrete particle swarm optimization(PDPSO) algorithm, overcomes the limitation of the implementation technique based on classical optimization problem solving. According to the load situation of itself and its neighbors, the cloud workflow engine reasonably decomposes the huge solution space of composite cloud service, and uses the Map/Reduce model to parallel the traditional particle swarm algorithm, and finds the optimal or suboptimal solution acceptable solution in polynomial time or near polynomial time.