Energy Optimization for Virtualized Network Environments

It appears your Web browser is not configured to display PDF files. Download adobe Acrobat or click here to download the PDF file.

Click here to download the PDF file.

Creator: 

Ghazisaeedi, Ebrahim

Date: 

2016

Abstract: 

Information and Communication Technology (ICT) has been estimated to consume 10% of the total energy consumption in industrial countries. According to the latest measurements, this amount is rapidly increasing by 6% annually. With the evolved new business model in which Service Providers (SPs) are separated from Infrastructure Providers (InPs), Virtualized Network Environments (VNEs) have been regarded as a promising technology for flexibly utilizing shared communication network resources. VNEs also play a fundamental role toward virtualizing data centers. In this thesis, we suggest different feasible solutions to optimize the energy consumption in a VNE. In this regard, first, we review the corresponding literature in regard to the architecture of a VNE, its performance modelling, several power models, and also existing energy-saving solutions for VNEs. We approach the objective of optimizing the energy consumption in a VNE by defining and solving two main problems. The first problem optimizes the energy consumption in a VNE during the off-peak period. This is feasible by reconfiguring the mapping of already embedded virtual networks for the off-peak time. This is planned in two smaller and simpler sub-problems with increasing the complexity and higher energy-saving levels. Our solutions enable the providers to adjust the level of the reconfiguration and accordingly control the possible traffic disruptions. In the second problem, we propose a novel energy-efficient embedding method that maps heterogeneous MapReduce-based virtual networks onto a heterogeneous data center physical network, energy-wise. We introduce a new incast problem that specifically may happen in Virtualized Data Centers (VDCs). The proposed embedding process also controls the incast queueing delay.

Subject: 

Engineering - Electronics and Electrical

Language: 

English

Publisher: 

Carleton University

Thesis Degree Name: 

Doctor of Philosophy: 
Ph.D.

Thesis Degree Level: 

Doctoral

Thesis Degree Discipline: 

Engineering, Electrical and Computer

Parent Collection: 

Theses and Dissertations

Items in CURVE are protected by copyright, with all rights reserved, unless otherwise indicated. They are made available with permission from the author(s).