Load Forecasting Based on Optimized Random Forest Algorithm in Cloud Environment

Load Forecasting Based on Optimized Random Forest Algorithm in Cloud Environment

To solve the problem of unbalanced resource load in cloud data center, a resource load forecasting method which is based on random forest model from the perspective of resource load forecasting is proposed in the paper. This method combines genetic algorithm with random forest algorithm to solve the problem that random forest algorithm can not determine the combination of parameter in order to obtain the optimum forecasting effect. The results of experiment show that compared with the super parametric method of random forest model, which is optimized by random search, the one optimized by genetic algorithm proposed in this paper has higher forecasting accuracy.