Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences

COLLEGE OF ENGINEERING

UC Berkeley

OPCA: Fault-Tolerant Routing and Load-Balancing

Sharad Agarwal, Chen-Nee Chuah and Randy H. Katz

EECS Department
University of California, Berkeley
Technical Report No. UCB/CSD-02-1210
2002

http://www.eecs.berkeley.edu/Pubs/TechRpts/2002/CSD-02-1210.pdf

The number of stub ASes that participate in the inter-domain BGP peering sessions has grown. Most of these ASes are multi-homed to multiple upstream providers. We believe that this behavior exhibits a widespread intent to achieve fault tolerant and load balanced connectivity to the Internet. However, BGP today offers route fail-over times as long as 15 minutes, and very limited control over incoming traffic across multiple wide area paths. We propose an policy control architecture, OPCA, that runs as an overlay network on top of BGP. OPCA allows an AS to make route change requests at other, remote ASes to achieve faster route fail-over and provide capabilities to control traffic entering the local AS. The proposed architecture and protocol will co-exist and interact with the existing routing infrastructure.


BibTeX citation:

@techreport{Agarwal:CSD-02-1210,
    Author = {Agarwal, Sharad and Chuah, Chen-Nee and Katz, Randy H.},
    Title = {OPCA: Fault-Tolerant Routing and Load-Balancing},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2002},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2002/6192.html},
    Number = {UCB/CSD-02-1210},
    Abstract = {The number of stub ASes that participate in the inter-domain BGP peering sessions has grown. Most of these ASes are multi-homed to multiple upstream providers. We believe that this behavior exhibits a widespread intent to achieve fault tolerant and load balanced connectivity to the Internet. However, BGP today offers route fail-over times as long as 15 minutes, and very limited control over incoming traffic across multiple wide area paths. We propose an policy control architecture, OPCA, that runs as an overlay network on top of BGP. OPCA allows an AS to make route change requests at other, remote ASes to achieve faster route fail-over and provide capabilities to control traffic entering the local AS. The proposed architecture and protocol will co-exist and interact with the existing routing infrastructure.}
}

EndNote citation:

%0 Report
%A Agarwal, Sharad
%A Chuah, Chen-Nee
%A Katz, Randy H.
%T OPCA: Fault-Tolerant Routing and Load-Balancing
%I EECS Department, University of California, Berkeley
%D 2002
%@ UCB/CSD-02-1210
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/2002/6192.html
%F Agarwal:CSD-02-1210