Backup Network Design For Random Link Failures And Network Vulnerability Assessment

Today, with the network technologies fast development and applications diversify,the requirements for network performances increase gradually. The issues on networksurvivability and vulnerability have been paid more attentions. The related works on theresource allocation for random link failures make redundancy effect. Therefore, thispaper will introduce a dedicated backup network solution. In the field of the networkvulnerability, this paper will associate it with geographical factors, so that it will bemore in line with the practice when suffering from destruction.In chapter two and three, this paper focus on the resource allocation for randomlink failures. Firstly, the paper introduces the concepts of robust optimization anddedicated backup network. Robust optimization is used to solve the uncertainparameters, which is more in line with the actual situation. And the dedicated backupnetwork is used to provide a low-capacity backup network for random link failures.Upon a link failure in the primary network, traffic is rerouted through a preplanned pathin the backup network. We introduce a novel approach for dealing with random linkfailures, in which probabilistic survivability guarantees are provided to limit capacityover-provisioning. We show that the optimal backup routing strategy in this respectdepends on the reliability of the primary network. Specifically, as primary links becomeless likely to fail, the optimal backup networks employ more resource sharing againstbackup paths. We apply results from the field of robust optimization to formulate an ILPfor the design and capacity provisioning of these backup networks.In chapter four, we research the network reliability problem with geographicallycorrelated failures. Communication networks are vulnerable to natural disasters, suchreal-world events happen in specific geographical locations and disrupt specific parts ofthe network. Therefore, the geographical layout of the network determines the impact ofsuch events on the network’s connectivity. In particular, we aim to identify the mostvulnerable parts of the network. In this paper, we model the disaster event as a linesegment or a disk and develop polynomial time algorithms that find a worst-case linesegment cut and a worst-case circular cut. Our novel approach provides a promising new direction for network design to avert geographical disasters.