EVT: Evolutionary Algorithm for VPN Tree Provisioning

A Virtual Private Network (VPN) is an overlay network defined over a public network infrastructure like the Internet. A VPN provides customers with a secure, low cost, and more flexible communication environment than leased lines. The allocation of bandwidth for VPNs to meet the requirements specified by customers is an important traffic engineering research issue. A VPN workload model called hose-model, first introduced in [1], was developed to provide customers with a flexible and convenient way of specifying the bandwidth requirements of VPN endpoints. The VPN endpoints are connected according to a tree topology. The general problem of computing a constrained VPN tree with optimum bandwidth allocation is a hard combinatorial optimization problem [2]. In this work, we propose an evolutionary algorithm called EVT to optimize the total bandwidth reserved on edges of the VPN tree. EVT is based on the Simulated Evolution meta-heuristic originally introduced in [3]. Experimental results with Waxman network graphs show that EVT performs significantly better than the Primal-Dual algorithm reported in [2] with respect to final solution quality.