Resource Scheduling Strategy Based On Ant Colony Algorithm Research Under The Cloud Environment

Cloud computing is a new type of computing model which is evolyed by distributed computing,grid computing and parallel computing. Mainly using virtualization technology.all kinds of resources in the cloud data center become resource pool providing management and external services. and form business model of the user "given what you need and counting depend on your quantity". These resources are transparent to the user, the user only needs to know the services provided by the cloud data center and choose the services they need, and they don’t need to know the specific detail during the execution of the tasks and the specific location. finally the cloud center will send the result back to the users. The business potential of the cloud computing is tremendous, and it has a profound im pact on the prospective operation mode of the IT and it has become a hot topics in the study of domestic and foreign enterprises and research institutions now.Due to the heterogeneous nature of cloud computing environments, as well as autonomy nature and dynamic nature also as cloud environment using virtualization technology. cloud computing environment is greatly different with the previous resource allocation in distributed computing, parallel computing and grid computing, the purpose of the resource scheduling in the cloud is that with the expansion of the size of the cloud data center and the increasing number of users and tasks, the virtual sesourse of the data center resources can meet user requests of Qos while the resourse can be used in a reasonable and efficient way.On the basic of a brief analysis of cloud computing and the research situation of resource scheduling in cloud environment, summing up the advantages and disadvantages of the existing resource scheduling strategy and improvement direction, and introduction of the basic concepts of resource scheduling strategy for cloud computing and cloud environments and technology system, this paper mainly contains the following three aspects of work:firstly, it will analyze and research the traditional ant colony algorithm existing problems including high time span, low load balance and single optimization target etc;Secondly:after showing the basic ideas、characteristics and design elements of ant colony algorithm and simulated annealing algorithm, a new algorithm ant colony simulated annealing algorithm (ACOSA) is designed for the most common programming framework cloud Map/Reduce, this algorithm combine simulated annealing algorithm and ant colony algorithm to minimize the scheduled time as the main target, and it specially introduce the degree of matching factor for the tasks to the resources and load balancing, and its main idea is to use the ant colony algorithm to get a set of tasks to resources the optimal solution, then the solution through simulated annealing algorithm to optimize and update pheromone paths, and finally get global optimal solution. Thirdly:In the thesis, this paper will present a detailed description of the cloud computing simulation platform CloudSim and recompile it to achieve ACOSA algorithm proposed in this paper. By comparison with the cloud resource scheduling strategy based on the original ant colony algorithm,the experiment verify the proposed scheduling policy with good performance in the time span and load balancing.