| Title |
Optimizing Adaptive QoS for Data Transfers in Software-Defined Networks |
| DOI |
https://doi.org/10.5573/ieie.2021.58.10.49 |
| Keywords |
Software-defined networks; Adaptive QoS; Mixed-integer linear programming-based algorithm; Dynamic path computation; Data transfer in wide-area networks |
| Abstract |
Recently more distributed scientific applications such as real-time visualization require hard QoS from network services. We define a new notion of fairness, Adaptive Hard QoS fairness for hard QoS flows. We develop a mixed-integer linear programming-based algorithm for dynamic path computation, which guarantees requirements of all QoS flows. Using numerical simulations and a real network topology (Internet2), we compare our algorithm's performance with that of Max-Min fairness algorithm, which schedules flows based on max-min fairness of their current demands regardless of QoS requirements. Our algorithm guarantees zero QoS violation whereas Max-Min fairness algorithm has up to 15% for some traffic patterns. |