Cloud Computing Resource Load Banlancing Study Based On Complex Network And Improved Ant Colony Algorithm

In recent years, because of high stability, easy maintenance, low cost and other factors, cloud computing is changing the network service model rapidly. Cloud computing service provider provide computing service, storage service and other services with different size and different way for companies, enterprises and individuals. Cloud computing condolidate storage resource, computing resource and software to achieve the integrated management and separation use of resource and provide more high-quality but lower cost services for cloud computing users. However, the resource in cloud environment is very large, and its scale will continuely increase by current trend, as well as the cloud computing platform of different cloud computing service providers is heterogeneous, the load unbalance is easily engendered in cloud computing environment. The load unbalance will not only affect users’ experience, but also will cause the waste of resource in cloud environment.This paper analysis the characteristics of cloud computing resource detailedly firstly. By comparison with the characteristics of complex, We found that they have same network features: non-linear, a large number of constituent elements, distribution, dynamic, open and emerge. This paper builds modelling of the cloud computing resource by complex network knowledge, analysis and discusses the cloud computing resource from a more intuitive way.In addition, we found that the ant colony algorithm has a very good fit with complex networks. The ant colony algorithm has high ability in solving NP-hard problem. And the cloud computing resource load balance mechanism can be seen as a NP-hard problem. In this paper, we combine the ant colony algorithm and cloud computing node network and carried out experiments and analysis on this basis.To overcome its disadvantages, this paper also improved the ant colony algorithm. We can see that the improved ant colony algorithm is able to better achieve load balance of resource by analysis and comparison based on experiments. Finally the degree of satisfaction of users and the cloud computing service providers’ interest are improved.