ATTENTION/WARNING - NE PAS DÉPOSER ICI/DO NOT SUBMIT HERE

Ceci est la version de TEST de DIAL.mem. Veuillez ne pas soumettre votre mémoire sur ce site mais bien à l'URL suivante: 'https://thesis.dial.uclouvain.be'.
This is the TEST version of DIAL.mem. Please use the following URL to submit your master thesis: 'https://thesis.dial.uclouvain.be'.
 

Making BGP aware of the data plane

(2024)

Files

Deprez_16801900_2024.pdf
  • Open access
  • Adobe PDF
  • 4.25 MB

Details

Supervisors
Faculty
Degree label
Abstract
The Border Gateway Protocol (BGP) is an inter-domain routing protocol that makes decisions based on information from the control plane. These decisions are used in the data plane to forward traffic to their destination. However, the control plane does not contain information about the current state of the data plane. Our objective is to explore the possibility and implications of adding knowledge from the data plane in BGP. This master thesis is divided into two main steps. First, we explore the possibility of modifying the behavior of BGP if we know information about the data plane. We used a planned administrative shutdown as a study case. We settled a sub-objective, which is to reduce the number of BGP withdrawals sent between Autonomous Systems (ASes) when a route planned to be shut down is removed. We modified the goBGP implementation. Our experiment demonstrates that despite some limitations, such as dropping traffic or delaying route advertisement, we can modify the BGP behavior. In the second step, we perform active measurements in the data plane and use the results in the BGP decision process. We created the Route Quality Measurement (RQM) system, a prototype that measures and quantifies the quality of a route in the data plane according to a metric. We added this metric into goBGP. Our experiments show that with our solution, routes with better quality are chosen as the best route instead of routes that would have been chosen for policy reasons. However, our prototype shows limitations in terms of capacity and precision when faced with a high load.