MPLS Traffic Engineering Recovery Mechanisms

This chapter discusses the Multi-Protocol Label Switching (MPLS) traffic engineering (TE) technology. It reviews the motivation for deploying such a technology in a data network and provides some examples that illustrate the mode of operation. One of the major challenges of network design is the traffic engineering, that is, how to route the traffic so that the network resources are efficiently used. Various network types like the public voice networks, ATM, the frame relay, and the internet protocol (IP) use the TE methods. The various motivations for deploying MPLS TE in a network include—bandwidth optimization, strict quality of service (QoS) guarantees, and fast recovery. The chapter also describes the different components of the recovery cycle with global default restoration because the different components of the recovery time are highly dependent of the network characteristics. The resulting recovery time may vary from a few milliseconds to hundreds of milliseconds. Testing the MPLS traffic reroute in a lab made of a few routers results in a very short convergence time. In contrast, a network with 1000 nodes links with high propagation delays, and hundreds of TE LSPs to reroute requires a much more significant amount of time to converge.