Method for Determining Recovery Order Against Intermittent Link Failures

Abstract

High reliability is required in networks, and it needs to build robust networks. In content delivery services in particular, service interruptions due to disconnection of communication paths between a server and nodes that receive the service must be avoided. Contents delivery services use a master server that contains original contents and edge-servers that hold copies of the contents. It needs that communication paths exist between the master server and the edge-servers even in the network failure. From this background, there is a lot of research on network design method to keep high reliability. On the other hand, recovery scheduling has not considered intermittent link failures, so that, if link failures occur, it was assumed that the failed links are recovered immediately. However, indeed, there is a possibility that network failures occur by an aftershock. In fact, in the Great East Japan Earthquake in 2011 and Kumamoto earthquake in 2016, road networks and electric networks were damaged by not only the main shock but also the aftershocks. In such a case, material resource and human resource are restricted, so that it is necessary to determine the recovery order of the failed links. In this paper, we aim to determine an appropriate recovery order of failed links against intermittent link failures. We formulate the optimization problem determining the recovery order and design a heuristic algorithm. In addition, we evaluate the performance of the algorithm by applying it to the topology of various actual networks.