Predicting Specification Violations During BGP Convergence

Abstract

Analyzing a network's behavior during convergence is challenging due to its highly non-deterministic nature. To address this, we developed BGPseer, the first analyzer that predicts specification violations during BGP convergence without network disruption. To do this both accurately and fast, BGPseer builds a probabilistic network timing model based on hardware measurements that allows to sample BGP message orderings, from which BGPseer estimates violation times. We implemented BGPseer by extending an open-source BGP simulator and show that it achieves 85--99% accuracy in estimating violation times in less than ten seconds.