Dynamic and Parallel Resource Allocation with Analytical Performance Estimation for Virtualized Environment

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.

Supplemental Files: 

Creator: 

Lu, Qiao

Date: 

2022

Abstract: 

In this thesis, we propose an integrated approach for estimating the performance of virtual network resource allocation and a Genetic Algorithm (GA) based mechanism for online dynamic resource allocation in virtualized environment. The integrated approach is empowered by a novel loss network model with Dynamic Routing And Random Topology (DRART), which is combined with some existing models to create a synergy across different levels through an effective recursive process. Numerical results show the proposed integrated approach can provide accurate predictions on the performances of general virtual network embedding algorithms. We propose a virtual link mapping solution, i.e., Segment Based Genetic Algorithm (SBGA), which provides new definitions for genes and chromosomes in the Genetic Algorithm. Our SBGA approach enables parallel processing for searching optimal allocations. Our theoretical analysis shows that the execution time of our approach can be reduced to logarithmic time. To map virtual nodes and links to physical ones in one stage, we further develop an approach, named as GAOne. Our proposed GAOne approach applies the two-color graph coloring in graph theory to guide the crossover process in the Genetic Algorithm for valid solutions. Our simulation results show that the proposed GAOne approach is fast and efficient for online resource allocation applications in virtualized environment.

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).