TY - GEN
T1 - Towards impactful routing research
T2 - 2013 ACM CoNEXT Student Workshop 2013
AU - Schlinker, Brandon
AU - Zarifis, Kyriakos
AU - Cunha, Italo
AU - Feamster, Nick
AU - Katz-Bassett, Ethan
AU - Yu, Minian
PY - 2013
Y1 - 2013
N2 - Interdomain routing has long been primed for an overhaul. Designed for a previous era and since dragged into today's networks through complex configurations, little progress has been made on aligning interdomain routing capabilities with network operator needs. Yet substantial growth has occurred in tangential areas such as datacenter networking. We believe that the barriers to performing interdomain routing research must be lowered to enable researchers and operators to better understand existing problems and evaluate potential improvements. To this end, we propose a testbed that enables the emulation of autonomous systems and their internal routing domains. The testbed allows these emulated networks to advertise routes and exchange traffic with the real Internet, exposing them to the Internet's complex policies and inconsistencies. Our testbed enables experiments to have complete control over the network topology and glue-logic between routing domains, including choice of IGP protocol, routing engine, and route redistribution. In addition, experiments can exchange routes and traffic with ISPs in multiple peering locations across the Internet, enabling experiments which emulate complex, geographically distributed networks.
AB - Interdomain routing has long been primed for an overhaul. Designed for a previous era and since dragged into today's networks through complex configurations, little progress has been made on aligning interdomain routing capabilities with network operator needs. Yet substantial growth has occurred in tangential areas such as datacenter networking. We believe that the barriers to performing interdomain routing research must be lowered to enable researchers and operators to better understand existing problems and evaluate potential improvements. To this end, we propose a testbed that enables the emulation of autonomous systems and their internal routing domains. The testbed allows these emulated networks to advertise routes and exchange traffic with the real Internet, exposing them to the Internet's complex policies and inconsistencies. Our testbed enables experiments to have complete control over the network topology and glue-logic between routing domains, including choice of IGP protocol, routing engine, and route redistribution. In addition, experiments can exchange routes and traffic with ISPs in multiple peering locations across the Internet, enabling experiments which emulate complex, geographically distributed networks.
KW - Interdomain routing
KW - Internet routing
KW - Network architecture
KW - Routing protocols
KW - Testbeds
UR - http://www.scopus.com/inward/record.url?scp=84893189361&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84893189361&partnerID=8YFLogxK
U2 - 10.1145/2537148.2537153
DO - 10.1145/2537148.2537153
M3 - Conference contribution
AN - SCOPUS:84893189361
SN - 9781450325752
T3 - CoNEXT Student Workhop 2013 - Proceedings of the 2013 Workshop
SP - 31
EP - 33
BT - CoNEXT Student Workhop 2013 - Proceedings of the 2013 Workshop
PB - Association for Computing Machinery
Y2 - 9 December 2013 through 9 December 2013
ER -