CPU Load Prediction Using Support Vector Regression and Kalman Smoother for Cloud

Rongdong Hu, Jingfei Jiang, Guangming Liu, Lixin Wang

Abstract: Cloud service provider's revenue is tightly related with the QoS and resources utilization. One optimal strategy is to timely allocate resources in a fine-grained mode according to the actual resources demand of applications. The necessary precondition of this approach is obtaining future load information in advance. CPU has a significant effect on the performance of applications. So CPU load prediction is one of the most important requirements in the Cloud resources provisioning. We propose a multi-step-ahead CPU load prediction method based on Support Vector Regression which is suitable for the dynamic characteristics of applications and the complex Cloud computing environment. Kalman smoothing technology is integrated to further reduce the prediction error. Real trace data were used to verify the prediction accuracy and stability of our method, comparing with AR, BPNN and standard SVR.