TY - GEN
T1 - A practical congestion control scheme for named data networking
AU - Schneider, Klaus
AU - Yi, Cheng
AU - Zhang, Beichuan
AU - Zhang, Lixia
N1 - Funding Information:
This work is partially supported by the National Science Foundation under Grant No. 1345142, 1345318, and 1551057. Any opinions, findings, and conclusions expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The authors would like to thank Luca Muscariello and Giovanna Carofiglio for helpful feedback on the evaluation and Alex Afanasyev for providing support regarding ndnSIM
Publisher Copyright:
© 2016 ACM.
PY - 2016/9/26
Y1 - 2016/9/26
N2 - Traditional congestion control mechanisms are designed for end-toend connections and do not fit the Named Data Networking (NDN) architecture, in which content can be retrieved from multiple sources and through multiple paths. To fully exploit the NDN architecture, a congestion control scheme must consider the effects of in-network caching, multipath forwarding, and multicast data delivery. Moreover, the solution must not assume known link bandwidths or Data packet sizes, as these assumptions may not hold for overlay links, wireless links, or applications with varying Data packet sizes. In this paper we propose PCON: a practical congestion control scheme to address the above issues. PCON detects congestion based on the CoDel AQM (by measuring packet queuing time), then signals it towards consumers by explicitly marking certain packets, so that downstream routers can divert traffic to alternative paths and consumers can reduce their Interest sending rates. Our simulations show that PCON's forwarding adaptation reaches a higher total throughput than existingwork while maintaining similarRTT fairness. Moreover, PCON can adapt to the changing capacity of IP tunnels and wireless links, conditions that other hop-by-hop schemes do not consider.
AB - Traditional congestion control mechanisms are designed for end-toend connections and do not fit the Named Data Networking (NDN) architecture, in which content can be retrieved from multiple sources and through multiple paths. To fully exploit the NDN architecture, a congestion control scheme must consider the effects of in-network caching, multipath forwarding, and multicast data delivery. Moreover, the solution must not assume known link bandwidths or Data packet sizes, as these assumptions may not hold for overlay links, wireless links, or applications with varying Data packet sizes. In this paper we propose PCON: a practical congestion control scheme to address the above issues. PCON detects congestion based on the CoDel AQM (by measuring packet queuing time), then signals it towards consumers by explicitly marking certain packets, so that downstream routers can divert traffic to alternative paths and consumers can reduce their Interest sending rates. Our simulations show that PCON's forwarding adaptation reaches a higher total throughput than existingwork while maintaining similarRTT fairness. Moreover, PCON can adapt to the changing capacity of IP tunnels and wireless links, conditions that other hop-by-hop schemes do not consider.
UR - http://www.scopus.com/inward/record.url?scp=84994138790&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84994138790&partnerID=8YFLogxK
U2 - 10.1145/2984356.2984369
DO - 10.1145/2984356.2984369
M3 - Conference contribution
AN - SCOPUS:84994138790
T3 - ACM-ICN 2016 - Proceedings of the 2016 3rd ACM Conference on Information-Centric Networking
SP - 21
EP - 30
BT - ACM-ICN 2016 - Proceedings of the 2016 3rd ACM Conference on Information-Centric Networking
PB - Association for Computing Machinery, Inc
T2 - 3rd ACM International Conference on Information-Centric Networking, ACM-ICN 2016
Y2 - 26 September 2016 through 28 September 2016
ER -