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'.
 

Parallelizing a BGP implementation

Foré, Nicolas
(2023)

Files

Fore_14571800_2013.pdf
  • UCLouvain restricted access
  • Adobe PDF
  • 1.47 MB

Details

Supervisors
Bonaventure, Olivier
Faculty
Ecole polytechnique de Louvain
Degree label
Master [120] : ingénieur civil en informatique, à finalité spécialisée
Abstract
Border Gateway Protocol (BGP) is one of the core protocol of Internet. The first definition of BGP was in 1989. At this time, CPU were single core and computing power was limited to this core. Most open source implementations are based on old code, and does not use our modern CPUs to its fullest. This is a problem, as the routing table of BGP routers keep growing each year. When starting a router, it is known that BGP will take more than a minute to receives the routes from its neighbors and run the BGP decision process. This high convergence time can lead to sub-optimal routing or packet losses. This thesis focuses on modifying the BGP implementation of Bird so multiple threads can work on the reception of BGP update message. This was done by modifying the data structure used to store routes and modifying the processing of BGP update message. This new implementation was then connected to other routers to check the convergence time. Though the new implementations did not decrease the convergence time, they can still be improved.