Improving Convergence Time of Routing Protocols

One of the main design goals of the Internet is robustness against failures. Normally, this is accomplished by redundancy and dynamic routing protocols that automatically adapt to failures: If a link is unavailable, data packets can generally be sent via alternative paths. An essential requirement for this is a fast mechanism for failure detection, since routing protocols can only start to reroute traffic around problems as soon as they get aware of them. This paper proposes a novel design of a generic failure detection service to be utilized by routing protocols that aims at dramatically decreasing the detection times of today’s mechanisms. After the introduction of the concept and its evaluation, the integration of the new failure detection service into BGP is described. Furthermore, an example for the adaption of a routing protocol using the proposed service is given.