Multi-Constrained Routing in Networks with Shared Risk Link Groups

In this work, we study a multi-constrained routing problem in networks with SRLGs where a path between a source and a destination is determined such that both the path cost and the weight of SRLGs to which the links of the path belong are bounded. The path cost is measured as the sum of cost of links on the path while the weight of SRLGs is calculated as the sum of the weight of individual SRLGs along the path. The solution to this problem can be used to find risk bounded and cost bounded path for a connection, or to design algorithms that find a pair of low cost SRLG-diverse paths between a source and a destination for survivable service provisioning. The multi-constrained routing problem is solved in two steps. First, an algorithm is devised that tries to find a least cost path in the network where the path cost is defined to be the combined cost of links and SRLG weights along the path. Second, an intelligent search algorithm that integrates the algorithm of the first step is designed to solve the multi-constrained routing problem to effectively find a path with the least total link cost while the total weight of the SRLGs along the path is bounded. The performances of the proposed algorithms are evaluated via extensive simulation and are compared with the solutions obtained by integer linear programming. Our simulation studies show that the proposed algorithms produce excellent results.