Electronic Thesis and Dissertation Repository

Degree

Master of Engineering Science

Program

Electrical and Computer Engineering

Supervisor

Abdallah Shami

Abstract

Efficient virtualization methodologies constitute the core of cloud computing data center implementation. Clients are attracted to the cloud model by the ability to scale available resources dynamically and the flexibility in payment options. However, performance hiccups can push them to return to the buy-and-maintain model. Virtualization plays a key role in the synchronous management of the thousands of servers along with clients' data residing on them. To achieve seamless virtualization, cloud providers require a system that performs the function of virtual network mapping. This includes receiving the cloud client requests and allocating computational and network resources in a way that guarantees the quality of service conditions for clients while maximizing the data center resource utilization and providers' revenue. In this thesis, we introduce a comprehensive system to solve the problem of virtual network mapping for a set of connection requests sent by cloud clients. Connections are collected in time intervals called windows. Subsequently, node mapping and link mapping are performed. Different window size selection schemes are introduced and evaluated. Three schemes to prioritize connections are used and their effect is assessed. Moreover, a technique dealing with connections spanning over more than a window is introduced. Simulation results show that the dynamic window size algorithm achieves cloud service providers objectives in terms of generated revenue, served connections ratio, resource utilization and computational overhead. In addition, experimental results show that handling spanning connections independently improves the results for the performance metrics measured.

Moreover, in a cloud infrastructure, handling all resources efficiently in their usage, management and energy consumption is challenging. We propose an energy efficient technique for embedding online virtual network requests in cloud data centers. The core focus of this study is to manage energy efficiently in cloud environment. A fixed windowing technique with spanning connections is used. Our algorithm, and a technique for randomly embedding nodes and links are also explained. The results clearly show that the algorithm used in this study generated better results in terms of energy consumption, served connections and revenue generation.

Share

COinS