Interdomain Ingress Traffic Engineering through Optimized AS-Path PrependingReportar como inadecuado


Interdomain Ingress Traffic Engineering through Optimized AS-Path Prepending


Interdomain Ingress Traffic Engineering through Optimized AS-Path Prepending - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

In Interdomain Ingress Traffic EngineeringINITE, a -target- Autonomous System AS aims to controlthe ingress link through which the traffic of one or moreupstream source networks flows to the target network or toits customers. Currently, there are few methodologies forsystematic INITE. In practice, ISPs often attempt to manipulate,mostly in a trial-and-error manner, the AS-Pathlength attribute of upstream routes through a simple techniqueknown as prepending or padding. In this paper,we focus on prepending and propose a polynomial-time algorithmreferred to as OPV that determines the optimalpadding for an upstream route at each ingress link of thetarget network. Specifically given a set of -elephant- sourcenetworks for a particular customer of the target network,and a set of maximum load constraints on the ingress linksof the latter, OPV determines the minimum padding at eachingress link so that the load constraints are met, when it isfeasible to do so. OPV requires as input an AS-Path lengthestimate from each source to each ingress link. We describehow to estimate this matrix, leveraging the BGP LookingGlass Servers that are abundant today for monitoring interdomainrouting. To deal with unavoidable inaccuraciesin the AS-Path length estimates, and also to compensatefor the generally unknown BGP tie-breaking process in upstreamnetworks, we develop a robust variation RPV ofthe OPV algorithm. We show that RPV manages to identifya padding vector that meets the given maximum load constraints,when it is feasible to do so, even in the presence ofinaccurate AS-Path lengths and unknown BGP tie-breakingbehavior.



College of Computing Technical Reports -



Autor: Gao, Ruomei - Dovrolis, Constantinos - Zegura, Ellen W. - -

Fuente: https://smartech.gatech.edu/







Documentos relacionados