Analysis of path stability within autonomous systems

Abstract

Interactive and multimedia applications depend on the stability of end-to-end paths on the Internet for predictable performance and good quality of service. On the other hand, network providers depend on multiple paths to ensure fault tolerance and use load balancing between these paths to enhance the overall network throughput. In this study, we analyze the path dynamics within service providers' networks using a month of measurement data from the RIPE Atlas platform which collects path traces between a source and a destination pair approximately every 15 minutes. Determining path segments belonging to each Autonomous System (AS), we explore ingress to egress paths of an AS in addition to source to destination paths within the AS. We observe that 13.2% of ASes contain routing loops within their network indicating mis-configuration of routers. Couple of ASes had over a million loops with over 100 routers causing loops in path traces. Additionally, 77.6% of end-to-end routes have at least two alternative AS paths with some end-to-end routes going through hundreds of different AS paths. Analyzing trace segments between ingresses and egresses of an AS, we realized more than a third of the backbone ASes have consistent routing between any node pair, but others implement load balancing or have inconsistent routing policies across their network.