Towards a Deep Belief Network-Based Cloud Resource Demanding Prediction

Abstract

Predicting resource demands in cloud computing environment is very important in order to make cloud system run optimally. The existing work falls short in conducting prediction in an satisfiable accuracy. In this paper, we propose to use Deep Belief Network(DBN)-based approach for cloud resource demanding prediction, which can capture high variances in cloud metric data without hand-crafting specified features. We have evaluated the proposed approach with The Google cluster trace released in 2011 to show the effectiveness in terms of accuracy. It shows that this DBN-based approach can predict the short term resource demands in a very accurate way, and long term prediction with acceptable accuracy.

DOI: 10.1109/UIC-ATC-ScalCom-CBDCom-IoP.2015.194

9 Figures and Tables

Cite this paper

@article{Zhang2015TowardsAD, title={Towards a Deep Belief Network-Based Cloud Resource Demanding Prediction}, author={Weishan Zhang and Pengcheng Duan}, journal={2015 IEEE 12th Intl Conf on Ubiquitous Intelligence and Computing and 2015 IEEE 12th Intl Conf on Autonomic and Trusted Computing and 2015 IEEE 15th Intl Conf on Scalable Computing and Communications and Its Associated Workshops (UIC-ATC-ScalCom)}, year={2015}, pages={1043-1048} }