الفهرس 
Chapter 1 IntroductionOnly 14 pages are availabe for public view
 |  | from | 145 |  |  |
| | | from | 145 | | |
Abstract
Nowadays, there is a continuous change and advancement in the field of Information Technology industry and this imposes a lot of challenges on the development of networks. This situation needs more intelligence and ability to leverage the data from connected and distributed environments such as classrooms to terminals for the sake of educational and work purposes. Some of the advances driving these needs are Artificial Intelligence, Virtualization and Internet of Things (IoT). Hence, there is a growing need for using Software defined network (SDN) paradigm that offers numerous benefits for wide-area networks such as promoting data transfer efficiency, improving application performance and reducing deployment costs. The present work proposes an accurate queuing system of packet forwarding to study traffic behavior and measure network performance through a vEdge router, testing the efficiency of the proposed SD-WAN model. It also tries to study the factors affecting the arrival of the packet and its average waiting time in queuing before entering service, together with calculating the average time of the whole system experimental measurements with the MATLAB package. Performance reflects our proposed queuing model offers a more precise approximation of Software defined wide area network (SD-WAN) controller system performance than existing ones. Additionally, the numerical analysis of the proposed queueing model shows that packet forwarding performance in SD-WAN greatly depends on utilization factor, where the higher the utilization in the system, the longer the response time for packet processing, and thus the delay increases. In addition to the previous model, second proposed model was built using the M/H_2/C queuing model helps to minimize problems of traditional network, testing the efficiency of the proposed SD-WAN model, as well as studying the factors affecting the arrival of the packet and its average waiting time in queuing before entering service and calculating the average time of the whole system, thus reducing message-delay. Also, we made a comparison between the two models to discover the optimal model for SD-WAN design. Finally, both models were emulated using emulated virtual environment (EVE-NG) emulator using some modern technologies of technologies such as automation, Virtualization (Network functions virtualization) and cloud.