A hybrid congestion control scheme for Named Data Networking via SDN

Abstract

AbstractAs the mainstream congestion control in Named Data Networking (NDN), the hybrid control mechanism integrates the advantages of consumer and router control mechanisms. However, The concurrent control by both consumers and routers may results in degraded data transmission performance. Therefore, we propose a novel hybrid congestion control scheme for NDN, called ndnSieve. By using the characteristics of Software Defined Network (SDN) that can achieve global network information acquisition as well as global resource scheduling, ndnSieve manages to invoke consumer and router control mechanisms based on different congestion states in the network respectively and achieves the synergy between two kinds of mechanisms. The results of experiment demonstrate that, compared with existing works, ndnSieve has the lowest average round-trip time, relatively high level of throughput and fairness, effectively improving the Quality of Service (QoS) of the network.