A network differentiation service method using DA-VNE algorithm

Abstract

To address the issue of limited network resources being unable to provide efficient differentiated service, a virtual network mapping algorithm based on delay awareness has been proposed, along with an intelligent network traffic classification and adjustment architecture. By paying attention to delay awareness, the low-latency requirements of different applications are realized, and a framework that can provide customized services according to different types of traffic is designed. The experimental results demonstrated that the proposed virtual mapping algorithm reduced node average latency by 23.61 % and 34.28 %, respectively, compared to the other two algorithms. Furthermore, it reduced link average latency by 35.01 % and 41.82 %, respectively. The success rate of this algorithm in 20 virtual network requests was 83.59 %, which was 3.17 % and 26.65 % higher than the other two algorithms. The results indicated that the delay-aware cross-domain virtual network mapping algorithm had significant advantages in the nodes, links, and overall latency of virtual network requests. The network-differentiated service method based on this algorithm can greatly satisfy the latency demands of different applications and alleviate network traffic congestion. The designed network architecture is feasible and reasonable in network traffic scheduling.