A NOVEL MECHANISM BASED ON ARTIFICIAL LOGICAL SPIDER WEB FOR REROUTING IN MPLS NETWORKS

Abstract

Multiprotocol label switching (MPLS) is a hybrid solution that combines the advantages of easy forwarding with the ability of guaranteeing quality-of-service (QoS). To deliver reliable service, MPLS requires traffic protection and recovery. Rerouting is one such recovery mechanism. In this paper, we propose a novel rerouting model called DDRAAS that is inspired by the spider and its web in nature. We try to establish an artificial logical spider web in the MPLS network to reorganise it into a structure that is more regular and simple. Based on this, we give the definition of the reroute area. Artificial spiders are then used to explore recovery paths dynamically in the reroute area. DDRAAS can be used to calculate the recovery paths in advance in order to protect the work path, while the improved DDRAAS can be a fast rerouting algorithm to calculate and establish recovery paths when faults occur. We have simulated our mechanism using the MPLS network simulator (MNS) and the performance metrics were compared to those of other proposals. The simulation results show that our mechanism is better in reducing packet loss, disorder and has faster rerouting speed. These improvements help to minimise the effects of link failure and/or congestion.