论文部分内容阅读
论述了云计算资源需求预测的作用,提出了新的基于夹角余弦的广义模糊软集相似性度量方法,将相似性度量结果与预测精度相结合来获得各单项预测模型的权重,并针对云计算环境中资源需求所表现出的短期动态性和长期周期性特征,选用自适应神经模糊推理系统ANFIS和季节性ARIMA模型SARIMA作为单项预测模型,以此构建基于广义模糊软集理论的组合预测模型GFSS-ANFIS/SARIMA。最后将该模型用于云计算环境下的资源需求预测应用中去。实验结果表明,与其它预测模型相比,该模型能有效提高预测精度,具有良好的预测性能。本文所提方法能为云计算资源的高效调度和分配提供决策支持。
This paper discusses the role of cloud computing resource demand forecasting and proposes a new method of similarity measure of generalized fuzzy soft sets based on cosine angle cosines. Combining the similarity measure results with the forecasting precision, we can obtain the weight of each individual forecasting model, The short-term dynamic and long-term periodicity characteristics of resource requirements in computing environment are calculated. The adaptive neuro-fuzzy inference system ANFIS and the seasonal ARIMA model SARIMA are selected as the single-item forecasting models to construct the combined forecasting model based on generalized fuzzy soft-set theory GFSS-ANFIS / SARIMA. Finally, the model is used in the prediction of resource demand under the cloud computing environment. Experimental results show that compared with other prediction models, this model can effectively improve the prediction accuracy and has good prediction performance. The proposed method can provide decision support for the efficient scheduling and allocation of cloud computing resources.