Performance Evaluation of a Distributed Scheme for Protection against Single and Double Faults for MPLS Fabio Ricciato, Marco Listanti, Angelo Belmonte, Daniele Perla INFOCOM dept.  University of Rome La Sapienza {fabio, listanti}@infocom.uniroma1.it Abstract. MPLS can be used to provide network robustness to faults through path protection techniques. In this paper we present a dynamic model support- ing different classes of end-to-end protection, including protection against Sin- gle Fault and Dual Fault, with and without sharing of backup bandwidth. Be- yond link and node failures we also consider protection against Shared Risk Link Group (SLRG) failure. An interesting feature of the proposed scheme is the ability to offer service differentiation with respect to the recovery probabil- ity, by coupling the differentiation on the number of backup paths with band- width assignment policy. In this paper we describe the underlying algorithms for route selection and backup bandwidth sharing. The route selection is based on explicit load-dependent routing of service and backup paths. We show by simulation that the proposed route selection algorithm is effective in improving the network utilization. We discuss two alternative implementations of our model: distributed and partially centralized. The primary concern with the dis- tributed approach is the message overhead implied by link-load dissemination, e.g. by flooding. However we show by simulation that message overhead can be taken under control by adopting a well-tuned adaptive overhead reduction algorithm. Our conclusion is that both distributed and partially-centralized im- plementation are feasible. 1 Introduction Telecommunication networks are facing today an evolutionary trend towards a dy- namic paradigm: more and more network functionalities are being shifted from the management plane - that implicitly foresees a sensible human intervention - towards the control plane  with a minimal human intervention. At the same time, the migra- tion of new critical application on IP networks is changing the service paradigm asso- ciated to such networks. From delivering simple best-effort service, IP networks are challenged today to offer bandwidth and reliability guarantees, along with service dif- ferentiation. Given the dimensions of current networks in size and bandwidth, each network operation model aimed at answering this challenge should be scalable first, then effective in resource usage. In this framework, MPLS and the related dynamic protocols constitute a useful add-on to traditional IP platforms. MPLS is currently be- ing implemented in several IP networks, and it can be exploited to implement connec- tion-oriented traffic engineering techniques. M. Ajmone Marsan et al. (Eds.): QoS-IP 2003, LNCS 2601, pp. 218-232, 2003.  Springer-Verlag Berlin Heidelberg 2003