Traffic Intensity-Based QoS Control in Software-Defined Network Test phase, srs, design phase and source code final deliverable

Traffic Intensity-Based QoS Control in Software-Defined Network Test phase, srs, design phase and source code final deliverable

Project Domain / Category

Computer Networks / Research / Simulation

Summary/Abstract

This project requires setting up NS3 and configuring the OpenFlow QoS controller as part of software-defined networking(SDN). Next, students must write the code to model a traffic intensity-based load balancing controller. It counts the number of TCP connections allocated to each server and measures the traffic intensity of each TCP session using the Erlang Formula every ten seconds. Finally, Students will be required to simulate a scenario that contains eight servers and 32 clients and perform the measurements in terms of TCP connection Latency, End to End Delay, Jitter, and Packet loss.

Functional Requirements:

  1. Environment preparation
    1. Install and configure a suitable version of the Ubuntu Operating system.
    2. Install and Configure a suitable version of NS3 that satisfies the requirements of OFSWITCH13.
    3. Install and intergerate the OFSWITCH13.
  2. Simulation Scenario
    1. Write suitable Simulation code to create the scenario given in Figure 1.
    2. Configure each server to accept TCP connections on port 5000. Each user should start TCP sessions of random duration between 5 to 15 seconds. User nodes should send TCP packets at random rates between 100kpbs to 200kpbs.

Figure 1 – Simulation Scenario

  1. Traffic Intensity Load Balancer
    1. Write a suitable load balancing algorithm to measure the traffic intensity at each server and allocate the incoming TCP sessions to a server having the lowest traffic intensity. The traffic intensity should be calculated as follows.

ℎ ∈ 15 ∗ ∗ ℎ

=

∗ ∗ ∗ 15

If all servers operate at the same traffic Intensity, allocate the incoming session to the server with the lowest ID.

  1. Performance Measurement
    1. Measure the performance of each TCP session in terms of Average packet loss, Connection Latency, delay, throughput, and jitter.

Tools:

Supervisor:

Name: Arif Husen

 

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top