Original Articles
Fangjin Zhu and Hua Wang
Abstract
Traditional Internet architecture is far too rigid for use with large numbers of network applications with different quality of service requirements. One new and promising approach to overcome the rigidity is network virtualization (NV), which allows multiple heterogeneous virtual networks to coexist on a shared substrate network (SN). However, one of the key problems for NV is the virtual network embedding (VNE) problem, which concerns the efficient mapping of virtual nodes and links to SN nodes and paths. The VNE problem has proven to be nondeterministic polynomial-time hard and approximation algorithms are needed to address it. In this paper, we define the VNE problem based on the multiple-choice knapsack model and propose a modified ant colony optimization algorithm to solve the problem. The combination of revenue and acceptance ratio of an SN is used as an important component when designing the fitness function to evaluate iterative solutions obtained by ants, and pheromone update rules are designed based on the fitness function. The cost of a candidate network is defined as the selection heuristic information. Simulation results show that this algorithm performs well with various numbers of VN requests. The algorithm also provides better optimization performance than existing algorithms.