Cloud task and virtual machine allocation strategy in cloud computing environment

Abstract

Targeted on cloud application services in cloud computing environment, a series of cloud task scheduling and virtual machine allocation strategies which are simple and easy to be implemented are put forward. There are five allocation strategies. They are the random allocation strategy, the full sequence allocation strategy, the sequence allocation strategy that the cloud tasks are sorted by the execution time before the tasks are assigned to the virtual machines by turns, the sequence allocation strategy that the virtual machines are sorted by the execution speed before the tasks are assigned to the virtual machines by turns and the greedy strategy that the load balancing is taken into account. The optimization objective is the total execution time of all tasks. Simulation and experimental analysis is operated on the simulation platform Cloudsim. Experimental results show that the time of the greedy strategy is the least, the time of the random strategy is the largest, and the time of the three sequence strategies are moderate. © Springer-Verlag Berlin Heidelberg 2012.