Network Performance Analysis of Carrier Grade Network Address Translation with Network Traffic Prioritization

No Thumbnail Available

Date

2021-11

Journal Title

Journal ISSN

Volume Title

Publisher

Addis Ababa University

Abstract

In recent years, Internet traffic increased very rapidly due to network technology growth. Additionally, the availability of fixed broadband is continually growing; such as those based on Asymmetric Digital Subscriber Line (ADSL) and fiber technologies. This huge demand is due to technological advancement, an increase in smartphone users, and the availability of new services by Internet service providers (ISPs). With this increasing demand for fixed broadband Internet service, improvement is needed from ISP to meet customer quality of service demand. Internet Protocol version 4 (IPv4) addresses are used to link network devices in a packet-switched network. All regional Internet registry pools of public IPv4 addresses have been depleted. With the depletion of public IPv4 addresses, ISPs are attempting to find an alternative solution to the problem. Several techniques have been proposed to address public IPv4 depletion issues. Carrier-Grade NAT (CGN) is one of the techniques utilized to solve the problem. Despite the fact that CGN solves public IPv4 address depletion by sharing a single routable IPv4 address for multiple private IPv4 addresses, it has significant flaws. CGN interferes with the smooth operation of several applications. It also breaks the Internet's end-to-end functioning premise. The CGN process is one of the causes of consumer Internet service delays, which leads to customer speed issues. The aim of this thesis is to examine the effect of CGN on network performance and to provide a solution for enhancing network performance. Also, examines the impact of the suggested solution which is adopting CGN in conjunction with network traffic prioritization on network performance. The evaluations were carried out on two scenarios: one with CGN operation and the other CGN combined with network traffic prioritization. For implementing and evaluating network performance of the scenarios, simulation tools such as Graphical Network Simulator-3 (GNS3), Ostinato traffic generator, and monitoring tool Paessler Router Traffic Grapher (PRTG) are employed. The performance of the two scenarios is analyzed using network metrics such as packet loss, latency, and throughput. The analysis results indicate that when CGN is used in conjunction with traffic prioritization, throughput improves by 19% on average, latency improves by 21%, and packet loss improves by 20%.

Description

Keywords

CGN, IPv4, Performance Metrics, Traffic Prioritization, Address Depletion

Citation