Virtual Machine Based on Genetic Algorithm Used in Time and Power Oriented Cloud Computing Task Scheduling

摘要

In cloud computing, task scheduling is a challenging problem in cloud data center, and there are many different kinds of task scheduling strategies. A good scheduling strategy can bring good effectiveness, where plenty of parameters should be regulated to achieve acceptable performance of cloud computing platform. In this work, combined elitist strategy, three parameters values oriented genetic algorithms are proposed. Specifically, a model built by Generalized Stochastic Petri Nets (GSPN) is introduced to describe the process of scheduling in cloud datacenter, and then the workflow of the algorithms is showed. After that, the effectiveness of the algorithms is found to be valid by the simulations on CloudSim.