TY - GEN
T1 - AggFlow
T2 - 1st ACM CoNEXT Workshop on Cloud-Assisted Networking, CAN 2016
AU - Yu, Bong Yeol
AU - Yang, Gyeongsik
AU - Lee, Kyungwoon
AU - Yoo, Chuck
N1 - Publisher Copyright:
© 2016 ACM.
PY - 2016/12/12
Y1 - 2016/12/12
N2 - In this paper, we propose AggFlow, a new address vir- tualization scheme for high scalability and resource effi- ciency. AggFlow is intended to provide complete ad- dress virtualization while incurring low overhead for physical switches and the control channel. To reduce overhead of address virtualization, we propose mapping- less address virtualization. We additionally introduce hop-by-hop-based forwarding, which aggregates ow rules installed in physical switches. This leads to efficient use of ternary content addressable memory and bandwidth of the control channel. Our evaluation using Mininet with simple linear and tree topologies, the ow table size of the core switch is reduced up to 1/16 for both topologies, and the control channel traffic decreases by 51% and 29% respectively. Furthermore, we measure the overhead incurred from AggFlow. Compared to OpenVirteX, AggFlow increases control plane delay by 0.1 ms, only 0.4% increase of CPU usage, and no change of throughput for UDP traffic.
AB - In this paper, we propose AggFlow, a new address vir- tualization scheme for high scalability and resource effi- ciency. AggFlow is intended to provide complete ad- dress virtualization while incurring low overhead for physical switches and the control channel. To reduce overhead of address virtualization, we propose mapping- less address virtualization. We additionally introduce hop-by-hop-based forwarding, which aggregates ow rules installed in physical switches. This leads to efficient use of ternary content addressable memory and bandwidth of the control channel. Our evaluation using Mininet with simple linear and tree topologies, the ow table size of the core switch is reduced up to 1/16 for both topologies, and the control channel traffic decreases by 51% and 29% respectively. Furthermore, we measure the overhead incurred from AggFlow. Compared to OpenVirteX, AggFlow increases control plane delay by 0.1 ms, only 0.4% increase of CPU usage, and no change of throughput for UDP traffic.
KW - Address virtualization
KW - Network virtualization
KW - Software defined networking
UR - http://www.scopus.com/inward/record.url?scp=85009730070&partnerID=8YFLogxK
U2 - 10.1145/3010079.3012012
DO - 10.1145/3010079.3012012
M3 - Conference contribution
AN - SCOPUS:85009730070
T3 - CAN 2016 - Proceedings of the 2016 ACM Workshop on Cloud-Assisted Networking, co-located with CoNEXT 2016
SP - 1
EP - 6
BT - CAN 2016 - Proceedings of the 2016 ACM Workshop on Cloud-Assisted Networking, co-located with CoNEXT 2016
PB - Association for Computing Machinery, Inc
Y2 - 12 December 2016
ER -