Multidomain Diagnosis of End-to-End Service Failures in Hierarchically Routed Networks

Abstract

Abstract. This paper investigates an approach to improving the scalability and feasibility of probabilistic fault localization in communication systems by exploiting the domain semantics of computer networks. The proposed technique divides the computational effort and system knowledge among multiple, hierarchically organized managers. Each manager performs fault localization in the domain it manages and requires only the knowledge of its own domain. Since failures propagate among domains, domain managers cooperate with each other to find a consensus explanation of the observed disorder. We show through simulation that the proposed approach increases the effectiveness of probabilistic diagnosis and makes it feasible in networks of considerable size 1 .

A Distributed Spatio-Temporal Event Correlation Protocol for Multi-Layer Virtual Networks

“…Pencolé and Cordier [8] have, for example, proposed a divide-andconquer principle for aggregating and merging subdiagnoses. Similarly, a distributed hierarchical end-to-end event correlation method is described by Steinder and Sethi [10].…”

Fault Propagation Model for Ad Hoc Networks

“…SIMULATION SETUP In this section we evaluate the two situations mentioned above in a faulty link detection scenario for ad hoc networks. We consider direct connections between network nodes (links) as possible "root-causes/faults" that are responsible for causing malfunctioning of the indirect connections (paths) in the network (being manifested as symptoms) [10]. Although there exist many reasons for the path failures, such as configuration issues, in our model only the problem of link failures is investigated.…”

“…In such a case, bidirectional connectivity is presented as two separate links while building the corresponding FPM. The FPM for the connectivity problem is a bipartite graph with parentless and childless nodes denoted as links and paths, respectively [10]. Each node of this graph is assumed to take only one of two states, "faulty" and "not faulty", and the fault propagation model is determined by the logical network topology.…”

An distributed multi-domain oriented fault diagnosis algorithm

“…Steinder and Sethi [10] proposed a multi-domain fault localization approach to end-to-end service failure diagnosis in hierarchically routed networks. This approach divides the computational effort and system knowledge among multiple hierarchically organized managers.…”