TY - GEN
T1 - PCC
T2 - 12th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2015
AU - Dong, Mo
AU - Li, Qingxi
AU - Zarchy, Doron
AU - Godfrey, P. Brighten
AU - Schapira, Michael
N1 - Funding Information:
We sincerely thank our shepherd Jon Howell and our reviewers for their valuable comments. We thank Srinivasan Keshav, Jeff Mogul, and Artur Makutunowicz for providing detailed and valuable comments at various stages of the project. We thank Google's QUIC team, including Ian Swett, for ongoing help with PCC's integration with the QUIC framework and Abdul Kabbani for helpful suggestions on software scalability. Finally, we gratefully acknowledge the support of NSF Grant 1149895, a Google Research Award, and an Alfred P. Sloan Research Fellowship. %blankline%
Publisher Copyright:
© 2015 by The USENIX Association. All Rights Reserved.
PY - 2015
Y1 - 2015
N2 - TCP and its variants have suffered from surprisingly poor performance for decades. We argue the TCP family has little hope of achieving consistent high performance due to a fundamental architectural deficiency: hardwiring packet-level events to control responses. We propose Performance-oriented Congestion Control (PCC), a new congestion control architecture in which each sender continuously observes the connection between its actions and empirically experienced performance, enabling it to consistently adopt actions that result in high performance. We prove that PCC converges to a stable and fair equilibrium. Across many real-world and challenging environments, PCC shows consistent and often 10× performance improvement, with better fairness and stability than TCP. PCC requires no router hardware support or new packet format.
AB - TCP and its variants have suffered from surprisingly poor performance for decades. We argue the TCP family has little hope of achieving consistent high performance due to a fundamental architectural deficiency: hardwiring packet-level events to control responses. We propose Performance-oriented Congestion Control (PCC), a new congestion control architecture in which each sender continuously observes the connection between its actions and empirically experienced performance, enabling it to consistently adopt actions that result in high performance. We prove that PCC converges to a stable and fair equilibrium. Across many real-world and challenging environments, PCC shows consistent and often 10× performance improvement, with better fairness and stability than TCP. PCC requires no router hardware support or new packet format.
UR - http://www.scopus.com/inward/record.url?scp=84967121175&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84967121175&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84967121175
T3 - Proceedings of the 12th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2015
SP - 395
EP - 408
BT - Proceedings of the 12th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2015
PB - USENIX
Y2 - 4 May 2015 through 6 May 2015
ER -