Path Restoration Versus Link Restoration in Survivable ATM Networks

Abstract

Network survivability is a critical issue in telecommunication networks due to the increasing dependence of the society on communication systems. Fast restoration from a network failure is an important challenge that deserves attention. This paper addresses an optimal link capacity and flow assignment design problem for survivable asynchronous transfer mode (ATM) networks based on two restoration strategies: a path restoration and a link restoration. Given the projected traffic demands and the network topology, the capacity and flow assignment are jointly optimized to yield the optimal capacity placement and flow allocation. The problem is formulated as a large-scale nonconvex nonlinear multi-commodity flow problem and is solved by a special augmented Lagrangian method (called separable augmented Lagrangian algorithm (SALA)). Several networks with diverse topological characteristics are used in the experiments to compare the two restoration strategies. Numerical results demonstrate that the link restoration strategy requires more capacity than the path restoration strategy when using both symmetric and asymmetric traffic. This strategy also increases the routing cost, the capacity installation cost, and the total network cost.