Controlling the impact of bgp policy changes on ip traffic

The Internet consists of nearly 12,000 autonomous systems ( AS’ ) that exchange routing information using the Border Ga tew y Protocol (BGP). The operators of each network need to have co ntrol over the flow of traffic through the AS. However, BGP does not facilitate common traffic engineering tasks, such as bal ancing load across multiple links to a neighboring AS or dire cting traffic to a different neighbor. Solving these problems is di fficult because the number of possible changes to routing pol icies is too large to exhaustively test all possibilities, some chan ges in routing policy can have an unpredictable effect on the flow of traffic, and the BGP decision process implemented by router v endors limits an operator’s control over path selection. In this paper, we demonstrate that it is possible to predictablymodel the changes in traffic flows in response to BGP policy cha nges, given that policies are adapted in a certain fashion . Based on analysis of routing tables and traffic measurement s from the AT&T backbone, we show that operators can control the scale o f the traffic engineering problem by focusing on the small fraction of destination prefixes (and sets of related prefixe s) r sponsible for the majority of traffic. Furthermore, the y can make the effects of their changes more predictable by following s pecific policy guidelines and selecting configuration optio ns that make the BGP decision process deterministic. This allows an operator to gain more control over network traffic within the existing BGP framework . Pages of Text 0 Other Pages 20 Total 20 No. Figs. 10 No. Tables 0 No. Refs. 27

[1]  V. Paxson.  End-to-end routing behavior in the internet , 2006, CCRV.

[2]  Anja Feldmann,et al.  NetScope: traffic engineering for IP networks , 2000, IEEE Netw..

[3]  G. Huston,et al.  Interconnection, Peering and Settlements , 2003 .

[4]  Larry Peterson,et al.  Inter-AS traffic patterns and their implications , 1999, Seamless Interconnection for Universal Services. Global Telecommunications Conference. GLOBECOM'99. (Cat. No.99CH37042).

[5]  Anja Feldmann,et al.  Deriving traffic demands for operational IP networks: methodology and experience , 2000, SIGCOMM.

[6]  Lixin Gao,et al.  Stable Internet routing without global coordination , 2000, SIGMETRICS '00.

[7]  Mikkel Thorup,et al.  Internet traffic engineering by optimizing OSPF weights , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[8]  Abhijit Bose,et al.  Delayed Internet routing convergence , 2000, SIGCOMM.

[9]  Christophe Diot,et al.  Understanding traffic dynamics at a backbone POP , 2001, SPIE ITCom.

[10]  D. O. Awduche,et al.  MPLS and traffic engineering in IP networks , 1999, IEEE Commun. Mag..

[11]  Abhijit Bose,et al.  Delayed internet routing convergence , 2000, SIGCOMM.

[12]  V. Monova.  Internet Routing Stability , 2003 .

[13]  Daniel O. Awduche,et al.  Requirements for Traffic Engineering Over MPLS , 1999, RFC.

[14]  Lionel M. Ni,et al.  Traffic engineering with MPLS in the Internet , 2000, IEEE Netw..

[15]  John W. Stewart,et al.  BGP4 : inter-domain routing in the Internet , 1998 .

[16]  Bassam Halabi,et al.  Internet Routing Architectures , 1997 .

[17]  kc claffy,et al.  Analysis of RouteViews BGP data: policy atoms , 2001 .