Analysis of link failures in an IP backbone

Abstract

Today's IP backbones are provisioned to provide excellent performance in terms of loss, delay and availability. However, performance degradation and service disruption are likely in the case of failure, such as fiber cuts, router crashes, etc. In this paper, we investigate the occurence of failures in Sprint's IP backbone and their potential impact on emerging services such as Voice-over-IP (VoIP). We first examine the frequency and duration of failure events derived from IS-IS routing updates collected from three different points in the Sprint IP backbone. We observe that link failures occur as part of everyday operation, and the majority of them are short-lived (less than 10 minutes). We also discuss various statistics such as the distribution of inter-failure time, distribution of link failure durations, etc. which are essential for constructing a realistic link failure model. Next, we present an analysis of routing and service reconvergence time during a controlled link failure scenario in our backbone. Our results indicate that disruption to packet forwarding after link failures depends not only on routing protocol dynamics, but also on the design of routers' architectures and control planes. Thus our resuits offer insights into two basic components for defining network-wide availability, which we consider a more appropriate metric for service-level agreements to support emerging applications.

Backtracking routing mechanism for improving link failure recovery

“…[2]. The suppression time after the link failure is detected is two seconds and the simulated time is one year.…”

“…The steady growth in the use of Internet for business-critical applications is the primary hidden reason [1] . It has been observed that majority of the link failures are short-lived and about half of the failures last less than a minute [2] . Hence, there is a growing demand for improving network survivability in the presence of short-lived link failures.…”

“…Traditional routing protocols such as open shortest path first (OSPF) [5] and intermediate system to intermediate system (ISIS) [6] deal with failures in a reactive manner, i.e. they rely on network wide link state advertisements to reroute around failures, taking seconds to resume forwarding [2,7,8] . To tradeoff between forwarding continuity and stability, fault insensitive routing (FIR) is presented in Ref.…”

Self-healing routing: failure, modeling and analysis

“…Markopoulou et al [9] have made a measurement study to understand the nature of intra-domain network failures recently, in which the study is carried out based on the early work proposed by Iannaccone et al [10]. They collected IS-IS routing updates from Sprint network and classified these updates into different categories based on the update root causes, such as maintenance-caused failures, router-related failures, optical-layer-related failures and individual-link failures.…”

Policy-constrained bio-inspired processes for autonomic route management

“…In this paper we discuss how policy-constrained bio-inspired algorithms and processes for routing can allow a network to gracefully react to transient link failures, which are described by Iannaccone et al [2] as the primary cause of service degradation in current networks. More specifically, we specify routing algorithms and processes based on biological principles and describe how they can be integrated within a policy-based network management framework for autonomic networking.…”