Queuing model based end-to-end performance evaluation for MPLS Virtual Private Networks

Abstract

Monitoring the end-to-end Quality-of-Service (QoS) is an important work for service providers' Operation Support System (OSS), because it is the fundamental requirement for QoS provisioning. However, it is in fact a challenging work, and there are few efficient approaches to address it. In this paper, for Multi-Protocol Label Switching (MPLS) Virtual Private Networks (VPNs), we propose a queuing model based end-to-end performance evaluation scheme for OSS to monitor the end-to-end delay, which is one of the important QoS metrics. By means of a queuing model, we deduce the relationship between the end-to-end delay and the information available in the Management Information Base (MIB) of routers, and then we present the evaluation scheme which avoids the costly per-packet measurement. The complexity of the proposed scheme is much lower than existing schemes. Extensive simulation results show that the proposed scheme can efficiently evaluate the end-to-end performance metrics (the estimation error is nearly 10%).