A Tractable Stochastic Model of Correlated Link Failures Caused by Disasters

Abstract

In order to evaluate the expected availability of a service, a network administrator should consider all possible failure scenarios under the specific service availability model stipulated in the corresponding service-level agreement. Given the increase in natural disasters and malicious attacks with geographically extensive impact, considering only independent single link failures is often insufficient. In this paper, we build a stochastic model of geographically correlated link failures caused by disasters, in order to estimate the hazards a network may be prone to, and to understand the complex correlation between possible link failures. With such a model, one can quickly extract information, such as the probability of an arbitrary set of links to fail simultaneously, the probability of two nodes to be disconnected, the probability of a path to survive a failure, etc. Furthermore, we introduce a pre-computation process, which enables us to succinctly represent the joint probability distribution of link failures. In particular, we generate, in polynomial time, a quasilinear-sized data structure, with which the joint failure probability of any set of links can be computed efficiently.

Network Resiliency Against Earthquakes

“…We have used the backbone topology of Interoute's network, which is the largest privately owned Europe-wide IP cloud. They also act as an infrastructure provider, and the locations of their fibers are shown on their web page 5 . We created a copy of their network in the Italian region using the yEd graph drawing tool.…”

“…J. Tapolcai et al presented a tractable stochastic model for regional link failures taking into account the correlation between the failures [5]. In case of any link set S, the output of the model tells the probability that when the next disaster happens, the set of failed links will be exactly S.…”

“…To model regional failures caused by an earthquake, we apply the findings of [4] and [5]. As in the present setting, both approaches translate to the same model, we will only present the common way of modeling earthquakes.…”

“…Sections III and V describe how to calculate these earthquake activity rates, in particular given the data available for Italy (namely, a catalog of earthquakes in the last 1000 years and intensity prediction equations, correlating the damage to the distance from the epicenter, which were based on observations within Italy). • We show, in Section IV, how to integrate earthquake activity rates into two existing network disaster resiliency models: (1) the stochastic geographically correlated link failures model in [5], and (2) the disaster set and failure states model from [4]. Although these models take a different approach towards computing the availability of network services under natural disasters, both approaches give exactly the same results when applied to our empirical earthquake model.…”

The Earth is nearly flat: Precise and approximate algorithms for detecting vulnerable regions of networks in the plane and on the sphere

“…An SRLG is called regional if it aims to characterize a failure damaging the network only in a bounded geographical area. It is still ongoing research on how to define and compute efficiently regional SRLG lists . A common simplification of these works is that they compute the list of SRLGs on a planar representation of the networks; thus, our focus is to generalize these approaches to the sphere.…”