Skip to Main Content
This paper describes a novel ultrafast heuristic algorithm to address an NP-hard optimization problem. One of the significant and nonintuitive results is that the schemes based on the heuristic can achieve a better overall performance than their time-consuming integer-linear-programming-based counterparts. More specifically, the proposed schemes are useful in providing an online shared path protection by establishing survivable connections in high-speed networks. The advantage of our heuristic algorithm over the existing heuristic algorithms in finding a pair of link-disjoint paths, which are called active and backup paths, comes from the following salient feature. It uses a so-called potential-backup-cost (PBC) function when selecting an active path in the first phase to take into consideration the backup bandwidth needed by the corresponding backup path, which has yet to be chosen in the second phase. The PBC function is derived mathematically based on the statistical analysis of experimental data. While the use of PBC only requires partial aggregate information on how the existing connections are established in a network, it can also be applied even more effectively when complete information is available.