| Cloud computing is a kind of dynamic business model which provides virtualized resources to user through the Internet. Data center is a very important part of the cloud computing. It hosts all of the resources, including computing resources, storage resources, networking and security resources and so on. Services are deployed in servers within the data center room. With the rapid development of cloud computing, the scale of data center also expands rapidly. On the one hand, the number of energy consumption from the hardware facilities becomes larger, on the other hand, the inefficient utilization of resources causes more waste.In order to solve the above problems, this paper focus on the optimizing strategy of virtual machine placement during the initialization of a data center. First of all, we determine that the optimizing goal is reducing the power consumption and network overhead in the entire data center on the basis of meeting customer service quality. Secondly, we propose a Modified Max-Min Ant Colony Optimization (M3 ACO) algorithm by modifying the pheromone updating rules and parameter settings. The algorithm can realize the global optimization, and there is a positive feedback mechanism, it can get the efficient optimal solutions by updating the pheromone. It was used for solving our problem in this paper. It is able to minimize power consumption and the network overhead, and get the best solutions on a high speed. Finally, we do lots of experiments to validate M3 ACO algorithm, and the experimental results show that the algorithm has availability, stability and expansibility.We apply the M3 ACO algorithm to the current cloud data center architecture system in this paper. We analyze the disadvantages of this architecture on the management, and then we apply the algorithm to the latest software defined data center architecture system. We analyzes its feasibility, convenience of the management and operation in the cloud data center. |
PDF Full Text Request