Environment data processing for a data centre

Maintaining a data centre is increasing in cost over the past decade due to the introduction of high computing performance. Two major energy consumers in the data centres are the computing systems and the cooling systems. The cooling systems are required to prevent the data servers from overheating...

Full description

Saved in:
Bibliographic Details
Main Author: Tan, Stella Pei Ting
Other Authors: Tan Rui
Format: Final Year Project
Language:English
Published: 2019
Subjects:
Online Access:http://hdl.handle.net/10356/77172
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Maintaining a data centre is increasing in cost over the past decade due to the introduction of high computing performance. Two major energy consumers in the data centres are the computing systems and the cooling systems. The cooling systems are required to prevent the data servers from overheating due to the high computing power. Hence, when looking at ways to save cost, one would investigate making the cooling systems more energy efficient while maintaining its intended functionalities. This project aims to look into making a specific part of the cooling system more energy efficient, and that is the cooling fan system. In order to achieve that, we first have to understand what are the factors that affects the energy consumption of the cooling fan system. This paper proposes that the environmental factors surrounding the data centre would affect the energy consumption of the cooling fan system, and thus it aims to prove the proposal through using data analytics techniques like multiple regression, Keras and KNN regression. The data is generated using various environmental sensors used in a research project in Nanyang Technological University. The findings of this paper indicate that environmental factors like the airflow speed, temperature and humidity do in fact affect the energy consumption of the cooling fan system, with the airflow speed being the most significant factor. By restricting the airflow speed into the room, one may reduce energy consumption by the cooling fan system and consequently reducing electricity cost. A predictive model of the energy consumption of the cooling fan system is also produced in order to help indicate whether a fan is consuming much more energy than a usual one, hence actions like replacing the fan can be taken to reduce electricity cost.