A Comparative Performance Evaluation of IPv4/IPv6 Using Network Simulation and Virtualization Tools

Abstract

Internet users are expected to grow from 4.5 billion in 2020 to 5.3 billion by 2023. In terms of the population, this represents 66% of the global population. However, the growth of connected devices and internet users is going very fast. Not yet as the number of the internet of things (IoT) applications such as video surveillance, smart meters, and tracking and the machine to machine increased as well. The address space of the IPv4 cannot meet these requirements and this growth in the connected devices. According to that another IP address scheme designed to meet the requirements and the needs for the IoT & M2M and the growth of the number of internet users. This is the IPv6 which consist of 128 bit and can provide a huge number of addresses. The limitation of the addressing space was solved by the IPv6 but the problem arises in the migration from IPv4 to IPv6 as the migration cannot happen overnight so transition mechanisms developed to allow the communication between both IPv4 and IPv6 network. This paper aims to study the performance parameters such as the delay, the throughput, and the TCP establishment time which could be affected by the transition mechanism. The methodology that used in this project is using a network simulation tools such as GNS3 and virtual box to simulate the network scenarios and network performance check such as Wireshark and iperf3 tools to analyse the traffic. The results in this project show that if the delay is the main concern, then the tunneling machine will be the best choice, behind that when the TCP flow playing an important role for the network engineer then the tunneling machine will be the best option. However, the dual-stack method is considered the best in terms of throughput.