A stochastic random-races algorithm for routing in MPLS traffic engineering

This paper1 presents an efficient adaptive online routing algorithm for the computation of bandwidth-guaranteed paths in Multiprotocol Label Switching (MPLS) based networks, using a learning scheme that computes an optimal ordering of routes. This work has two-fold contributions. The first is that we propose a new class of solutions other than those available in the literature incorporating the family of stochastic Random-Races (RR) algorithms. The most popular previously-proposed MPLS Traffic Engineering (TE) solutions attempt to find a superior path to route an incoming path setup request. Our algorithm, on the other hand, tries to learn an optimal ordering of the paths through which requests can be routed according to the "rank" of the paths in the order learnt by the algorithm. The second contribution of our work is that we have proposed a routing algorithm that has better performance than the important algorithms in the literature. The efficiency of our algorithm was experimentally established