Project H2DQ

Traditionally, for maintaining Quality of Service (QoS) proportional fairness and max-min fairness policy services are used independently. In this project, we propose an optimisation followed by two-dimensional queue management policy, called Hierarchical Two Dimension Queue (H2DQ) integrating both proportional fairness and max-min fairness. Here proportional fairness criterion decides the traffic shaping parameters for ensuring QoS among different traffic classes, and max-min fairness criterion determines the actual routing paths and shaping policies for different traffic flows considering the service level fairness.

H2DQ has been implemented and tested over a network setup emulated by using MiniNet emulator platform. We are extending this work by implementing H2DQ over CloudLab to study the effect of internet latency on our management policy.

Bhaumik, Sumitro & Chakraborty, Sandip. (2018). Hierarchical Two Dimensional Queuing: A Scalable Approach for Traffic Shaping using Software Defined Networking. 150-158. 10.1109/NETSOFT.2018.8460072.