TY - GEN
T1 - Recursively cautious congestion control
AU - Mittal, Radhika
AU - Sherry, Justine
AU - Ratnasamy, Sylvia
AU - Shenker, Scott
N1 - Funding Information:
We would like to thank all our colleagues in UC Berkeley, for their help and feedback - in particular Sangjin Han, Jon Kuroda, David Zats, Aurojit Panda and Gau-tam Kumar. We are also very thankful to our anonymous Hotnets 2013 and NSDI 2014 reviewers for their helpful comments and to our shepherd Prof. Srinivasan Seshan for his guidance in shaping the final version of the paper. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant No. DGE-1106400.
Funding Information:
We would like to thank all our colleagues in UC Berkeley, for their help and feedback - in particular Sangjin Han, Jon Kuroda, David Zats, Aurojit Panda and Gautam Kumar. We are also very thankful to our anonymous Hotnets 2013 and NSDI 2014 reviewers for their helpful comments and to our shepherd Prof. Srinivasan Seshan for his guidance in shaping the final version of the paper. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant No. DGE-1106400.
PY - 2014
Y1 - 2014
N2 - TCP’s congestion control is deliberately cautious, avoiding network overloads by starting with a small initial window and then iteratively ramping up. As a result, it often takes flows several round-trip times to fully utilize the available bandwidth. In this paper we propose RC3, a technique to quickly take advantage of available capacity from the very first RTT. RC3 uses several levels of lower priority service and a modified TCP behavior to achieve near-optimal throughputs while preserving TCP-friendliness and fairness. We implement RC3 in the Linux kernel and in NS-3. In common wide-area scenarios, RC3 results in over 40% reduction in average flow completion times, with strongest improvements – more than 70% reduction in flow completion time – seen in medium to large sized (100KB - 3MB) flows.
AB - TCP’s congestion control is deliberately cautious, avoiding network overloads by starting with a small initial window and then iteratively ramping up. As a result, it often takes flows several round-trip times to fully utilize the available bandwidth. In this paper we propose RC3, a technique to quickly take advantage of available capacity from the very first RTT. RC3 uses several levels of lower priority service and a modified TCP behavior to achieve near-optimal throughputs while preserving TCP-friendliness and fairness. We implement RC3 in the Linux kernel and in NS-3. In common wide-area scenarios, RC3 results in over 40% reduction in average flow completion times, with strongest improvements – more than 70% reduction in flow completion time – seen in medium to large sized (100KB - 3MB) flows.
UR - http://www.scopus.com/inward/record.url?scp=85076933171&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076933171&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85076933171
T3 - Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2014
SP - 373
EP - 385
BT - Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2014
PB - USENIX Association
T2 - 11th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2014
Y2 - 2 April 2014 through 4 April 2014
ER -