|
Title:
|
PREDICTING RESOURCE REQUIREMENTS OF THE SERVERS INTO CLOUD DATACENTERS USING NEURAL NETWORKS |
|
Author(s):
|
Fréjus Gbaguidi, Selma Boumerdassi, Eugène Ezin |
|
ISBN:
|
978-989-8533-60-9 |
|
Editors:
|
Piet Kommers and Pedro Isaías |
|
Year:
|
2017 |
|
Edition:
|
Single |
|
Keywords:
|
DataCenter, Cloud, Prediction, Energy consumption |
|
Type:
|
Full Paper |
|
First Page:
|
129 |
|
Last Page:
|
136 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
As a result of the rise of the digital, the world is facing the unbridled growth in the energy consumption of Datacenters, which are still proliferate since recent years. Many techniques are developed but they fail to meet all Datacenters contexts because of they variation in terms not only of the hardware environment but also the volume of data processed. In this article, we try to evaluate the contribution of prediction based on neural networks in the resolution of this question. To do this, we set up different network architectures in which we inject different time scale data sets in order to find the best parameters for obtaining reliable and fast predictions. These predictions could be used later to improve the placement of virtual machines on servers or to plan virtual machine migrations to optimize the use of server resources within Datacenters. |
|
|
|
|
|
|
