Research on Energy Consumption Modeling Method for Multi-Type Hybrid Applications

DOI

10.2991/ammsa-17.2017.4How to use a DOI?

Keywords

cloud computing; power modeling; hybrid applications

Abstract

With the rapid development of cloud computing, a large number of datacenters were built. These datacenters provide excellent computing resource and service to people. However, they also bring huge energy consumption at the same time. Scientists all around the world try to solve this problem, for which the first step is to build an accurate power consumption model. Some related work aims to establish a model which only can be applied to a single-type application. Nevertheless, multi-type applications are often executed simultaneously in practice, and it is difficult to establish a multi-type application power consumption model. In this paper, we design a multi-type application power consumption model based on the hardware resource utilization ratio, recording the server power at different hardware resource utilization ratio. Using hardware resource utilization ratio as input and the power as output, we build the model by leveraging the BP neural network method. Finally, the accuracy of our method can reach 95%.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Fengyu Guo
AU  - Xiaoying Wang
PY  - 2017/05
DA  - 2017/05
TI  - Research on Energy Consumption Modeling Method for Multi-Type Hybrid Applications
BT  - Proceedings of the 2017 International Conference on Applied Mathematics, Modelling and Statistics Application (AMMSA 2017)
PB  - Atlantis Press
SP  - 14
EP  - 20
SN  - 1951-6851
UR  - https://doi.org/10.2991/ammsa-17.2017.4
DO  - 10.2991/ammsa-17.2017.4
ID  - Guo2017/05
ER  -