TY - GEN
T1 - SCL
T2 - 14th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2017
AU - Panda, Aurojit
AU - Zheng, Wenting
AU - Hu, Xiaohe
AU - Krishnamurthy, Arvind
AU - Shenker, Scott
N1 - Funding Information:
9 Acknowledgment We thank Colin Scott, Amin Tootoonchian, and Barath Raghavan for the many discussions that shaped the ideas in this paper; we thank Shivaram Venkataraman, our shepherds Minlan Yu and Jay Lorch, and the anonymous reviewers for their helpful comments. This work was funded in part by a grant from Intel Corporation, and by NSF awards 1420064 and 1616774.
Funding Information:
We thank Colin Scott, Amin Tootoonchian, and Barath Raghavan for the many discussions that shaped the ideas in this paper; we thank Shivaram Venkataraman, our shepherds Minlan Yu and Jay Lorch, and the anonymous reviewers for their helpful comments. This work was funded in part by a grant from Intel Corporation, and by NSF awards 1420064 and 1616774.
PY - 2017
Y1 - 2017
N2 - We consider the following question: what consistency model is appropriate for coordinating the actions of a replicated set of SDN controllers? We first argue that the conventional requirement of strong consistency, typically achieved through the use of Paxos or other consensus algorithms, is conceptually unnecessary to handle unplanned network updates. We present an alternate approach, based on the weaker notion of eventual correctness, and describe the design of a simple coordination layer (SCL) that can seamlessly turn a set of single-image SDN controllers (that obey certain properties) into a distributed SDN system that achieves this goal (whereas traditional consensus mechanisms do not). We then show through analysis and simulation that our approach provides faster responses to network events. While our primary focus is on handling unplanned network updates, our coordination layer also handles policy updates and other situations where consistency is warranted. Thus, contrary to the prevailing wisdom, we argue that distributed SDN control planes need only be slightly more complicated than single-image controllers.
AB - We consider the following question: what consistency model is appropriate for coordinating the actions of a replicated set of SDN controllers? We first argue that the conventional requirement of strong consistency, typically achieved through the use of Paxos or other consensus algorithms, is conceptually unnecessary to handle unplanned network updates. We present an alternate approach, based on the weaker notion of eventual correctness, and describe the design of a simple coordination layer (SCL) that can seamlessly turn a set of single-image SDN controllers (that obey certain properties) into a distributed SDN system that achieves this goal (whereas traditional consensus mechanisms do not). We then show through analysis and simulation that our approach provides faster responses to network events. While our primary focus is on handling unplanned network updates, our coordination layer also handles policy updates and other situations where consistency is warranted. Thus, contrary to the prevailing wisdom, we argue that distributed SDN control planes need only be slightly more complicated than single-image controllers.
UR - http://www.scopus.com/inward/record.url?scp=85076884091&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076884091&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85076884091
T3 - Proceedings of the 14th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2017
SP - 329
EP - 345
BT - Proceedings of the 14th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2017
PB - USENIX Association
Y2 - 27 March 2017 through 29 March 2017
ER -