Author

Melvin Ver

Abstract

Live migration of virtual machines (VMs) is the process of transitioning a VM from one virtual machine monitor (VMM) to another without halting the guest operating system, often between distinct physical machines, has opened new opportunities in computing. It allows a clean separation between hardware and software, and facilitates fault management, load balancing, and low-level system maintenance. Implemented by several existing virtualization products, live migration also aids in aspects such as high availability services, transparent mobility and consolidated management. While virtualization and live migration enable important new functionality, the combination introduces novel security challenges. A virtual machine monitor that incorporates a vulnerable implementation of live migration functionality may expose both the guest and host operating system to attack and result in a compromise of integrity. Given the large and increasing market for virtualization technology, a comprehensive understanding of virtual machine migration security is essential. So the main idea behind this thesis is to create a test environment that is suitable for experimenting and analyzing the security implications in case of exploitation of Live Migration of Virtual Machines. Using Live VM migration for dynamic load balancing or scheduling, this study determines workload hotspots in physical environment and through use of effective Live Migration process; tries to carry out resource profiling. By carrying out effective profiling, this thesis research is able to determine how much of each resource needs to be allocated to a VM. To understand exactly why process migration would not work in such scenarios and better understand Live VM Migration, this thesis tries to provide requisite incites as to which model is most appropriate for automatic load balancing for virtual machine infrastructure based on resource consumption. The security implications of exploiting the process of migration may end in unexpected results or results that are not noticeable. The scope of this thesis research is identifying these results and the causes for them.

Library of Congress Subject Headings

Virtual computer systems--Security measures

Publication Date

2011

Document Type

Thesis

Student Type

Graduate

Degree Name

Networking and System Administration (MS)

Department, Program, or Center

Computer Science (GCCIS)

Advisor

Border, Charles

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works in December 2013.

Campus

RIT – Main Campus

Share

COinS