Comparative Study on the Performance of Different TCP Flavors

• Main Author: Oumer, Abduaziz Jama
• Format: Thesis
• Language: English; English
• Published: 2009
• Subjects: TK5101-6720 Telecommunication
• Online Access: https://etd.uum.edu.my/1618/1/Abdulaziz_Jama_Omar_Abdi.pdf; https://etd.uum.edu.my/1618/2/1.ABDULAZIZ_JAMA_OMAR_ABDI.pdf; https://etd.uum.edu.my/1618/

Indeed the Transmission Control Protocol (TCP) is the main transport layer protocol for the end-to-end control that helps the creation of information communication. Most of today's Internet applications depend on the Performance TCP simply because the most frequently used networks by today are the TCP/IP networks. TCP was originally created to handle the problem of network congestion collapse. In this research project, we had investigated the performance of four TCP variants namely Reno, Vegas, NewReno and SACK based on two performance measures: The Bandwidth (effective throughput) and fairness. The network topology is simple wired network and it will be configured into different scenarios to maximize the chances of achieving the desired goal. Simulation methodology is used in this study. The simulation tool or software that was used as an investigation environment is the popular NS-2 simulator. The objective was to investigate and find out the performance of TCP variants according to the bandwidth and fairness in a simple dumbbell wired network, in a hope to observe a better performance. However, the results are daunting, TCP Reno is the most aggressive (least fair one), and highest amount of throughput. In the case of TCP NewReno it follows Reno's steps by becoming the second most aggressive(second least fair),and second highest throughput. SACK (Sackl) is fair to Reno and NewReno, but when it is competing with Vegas, it shows that it is very unfair. Finally Vegas shows the highest degree of fairness (least aggressive) and as well Vegas produces the lowest amount throughput.