A Data Placement Strategy Based on Genetic Algorithm in Cloud Computing Platform

Since cloud computing platform can provide infinite storage capacity, computing ability as well as information services, it now has become the popular new application platform for both individuals and enterprises. The storage capacity of a data center is limited. Therefore, how to place data slices in appropriate data center proves to be an important factor influencing the platform ability. The data placement strategy we design in this paper takes the cooperation costs among data slices into account. It lowers the distributed transaction costs as much as possible, especially the cost differences among different distributed transactions. At the same time, this strategy also cares about the global load balance problem in data center. It is developed on the basis of genetic algorithm and ensures that the strategy can quickly converge to efficient data placement solutions. According to the result of the experiment, this strategy can better realize the global load balance and can save about 10% of the distributed cooperation costs when being compared with other strategies.