Home // ENERGY 2016, The Sixth International Conference on Smart Grids, Green Communications and IT Energy-aware Technologies // View article

Optimization of Power Usage Effectiveness for Heterogenous Modular Data Centers using Neural Network

Authors:
Vishal Kumar Singh
Jinhua Guo

Keywords: Machine learning; Neural Network; PUE; Data center.

Abstract:
With the rise of Internet of Things (IoT), it is becoming cheaper and easier to collect data from data center (DC) mechanical, electrical and control systems. These systems have complex interactions with each other. The static control logics and high number of configuration and nonlinear interdependency create challenges in understanding and optimizing energy efficiency. This is particularly challenging and expensive in medium size or smaller configurations like data suites or modular data centers. We utilize a learning engine that learns from operationally collected data to accurately predict power usage effectiveness (PUE) and create a control model to validate test results. Using the machine learning framework developed in this paper, we are able to predict DC PUE within 0.0004 +/¬ 0.0005. The results show that machine learning can improve data suite efficiency. The results also indicate that neural network based controller shows promise for practical implementation.

Pages: 27 to 32

Copyright: Copyright (c) IARIA, 2016

Publication date: June 26, 2016

Published in: conference

ISSN: 2308-412X

ISBN: 978-1-61208-484-8

Location: Lisbon, Portugal

Dates: from June 26, 2016 to June 30, 2016