Joint QoS and congestion control based on traffic prediction in SDN

Abstract

Due to the various network requirements of applications, quality of service (QoS)-aware routing plays an important role in the networks. Recently proposed resource allocation algorithms focus on the current traffic matrix, which is not applicable for dynamic networks. In this paper, we exploit an estimation of flow matrix that gives our scheme the ability to sufficiently reduce the total packet loss and simultaneously raise the network throughput. In this way, we mathematically formulate the QoS-aware resource reallocation in software-defined networking (SDN) networks based on the traffic prediction. To solve this optimization problem, two schemes are proposed: (i) exact solution; and (ii) fast suboptimal one. The proposed schemes are compared with the accuracy perspective. Moreover, the impact of prediction on resource reallocation is discussed. In this regard, it is shown that, compared with the conventional scheme, the proposed scheme decreases the packet loss and increases the throughput significantly.