TY - GEN
T1 - Towards decentralized fast consistent updates
AU - Nguyen, Thanh Dang
AU - Chiesa, Marco
AU - Canini, Marco
N1 - Publisher Copyright:
© 2016 Copyright held by the owner/author(s).
PY - 2016/7/16
Y1 - 2016/7/16
N2 - Updating data plane state to adapt to dynamic conditions is a fundamental network control operation. Software-Defined Networking (SDN) offers abstractions for updating network state while preserving consistency properties. However, realizing these abstractions in a purely centralized fashion is ineficient, due to the inherent delays between switches and the SDN controller, we argue for delegating the responsibility of coordinated updates to the switches. To make our case, we propose ez-Segway, a mechanism that enables decentralized network updates while preventing forwarding anomalies and avoiding link congestion. In our architecture, the controller is only responsible for computing the intended network configuration. This information is distributed to the switches, which use partial knowledge and direct message passing to efficiently schedule and implement the update. This separation of concerns has the key benefit of improving update performance as the communication and computation bottlenecks at the controller are removed. Our extensive simulations show update speedups up to 2x.
AB - Updating data plane state to adapt to dynamic conditions is a fundamental network control operation. Software-Defined Networking (SDN) offers abstractions for updating network state while preserving consistency properties. However, realizing these abstractions in a purely centralized fashion is ineficient, due to the inherent delays between switches and the SDN controller, we argue for delegating the responsibility of coordinated updates to the switches. To make our case, we propose ez-Segway, a mechanism that enables decentralized network updates while preventing forwarding anomalies and avoiding link congestion. In our architecture, the controller is only responsible for computing the intended network configuration. This information is distributed to the switches, which use partial knowledge and direct message passing to efficiently schedule and implement the update. This separation of concerns has the key benefit of improving update performance as the communication and computation bottlenecks at the controller are removed. Our extensive simulations show update speedups up to 2x.
KW - Decentralized network update
KW - Software-defined networking
UR - http://www.scopus.com/inward/record.url?scp=84983372975&partnerID=8YFLogxK
U2 - 10.1145/2959424.2959435
DO - 10.1145/2959424.2959435
M3 - Conference contribution
AN - SCOPUS:84983372975
T3 - ANRW 2016 - Proceedings of the ACM, IRTF and ISOC Applied Networking Research Workshop
SP - 19
EP - 25
BT - ANRW 2016 - Proceedings of the ACM, IRTF and ISOC Applied Networking Research Workshop
PB - Association for Computing Machinery, Inc
T2 - ACM, IRTF and ISOC Applied Networking Research Workshop, ANRW 2016
Y2 - 16 July 2016
ER -