Optimizing BitTorrent-like peer-to-peer systems in the presence of network address translation devices

BitTorrent nowadays is one of the most important peer-to-peer (P2P) file-sharing applications on the Internet, and it has also inspired many other P2P applications such as live or on-demand video streaming services. On the other hand, Network Address Translation (NAT) has become pervasive in almost all networking scenarios, from residential Internet access to enterprise networks. Despite the effort of NAT traversal, it is still very likely that P2P applications cannot receive incoming connection requests properly if they are behind NAT. To quantify the performance impact of NAT on BitTorrent-like P2P systems, we have created and validated a detailed but tractable mathematical model. We have also proposed and briefly examined a simple but effective strategy to mitigate the negative impact on NAT peers. In this paper, we further extend our model to cover the proposed biased optimistic unchoke strategy, and optimize the system performance in terms of both average peer download time and system finish time. We also create a new, faster Java-based BitTorrent simulator, which is used to validate the extended model and show the optimality of the proposed strategy and its boundary conditions.

[1]  Johan A. Pouwelse,et al.  The Bittorrent P2P File-Sharing System: Measurements and Analysis , 2005, IPTPS.

[2]  Bin Fan,et al.  The Delicate Tradeoffs in BitTorrent-like File Sharing Protocol Design , 2006, Proceedings of the 2006 IEEE International Conference on Network Protocols.

[3]  Bo Li,et al.  DONet: A Data-Driven Overlay Network For Efficient Live Media Streaming , 2004, INFOCOM 2005.

[4]  Rayadurgam Srikant,et al.  Modeling and performance analysis of BitTorrent-like peer-to-peer networks , 2004, SIGCOMM 2004.

[5]  B. Cohen,et al.  Incentives Build Robustness in Bit-Torrent , 2003 .

[6]  Venkata N. Padmanabhan,et al.  Analyzing and Improving BitTorrent Performance , 2005 .

[7]  Johan A. Pouwelse,et al.  Modeling and analyzing the effects of firewalls and NATs in P2P swarming systems , 2010, 2010 IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum (IPDPSW).

[8]  Vishal Misra,et al.  BitTorrent: An Extensible Heterogeneous Model , 2009, IEEE INFOCOM 2009.

[9]  Johan A. Pouwelse,et al.  Free-Riding, Fairness, and Firewalls in P2P File-Sharing , 2008, 2008 Eighth International Conference on Peer-to-Peer Computing.

[10]  Jonathan D. Rosenberg,et al.  Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols , 2010, RFC.

[11]  Eddie Kohler,et al.  Clustering and sharing incentives in BitTorrent systems , 2006, SIGMETRICS '07.

[12]  Jianping Pan,et al.  The impact of NAT on BitTorrent-like P2P systems , 2009, 2009 IEEE Ninth International Conference on Peer-to-Peer Computing.

[13]  Pyda Srisuresh,et al.  Traditional IP Network Address Translator (Traditional NAT) , 2001, RFC.

[14]  Venkata N. Padmanabhan,et al.  Analyzing and Improving a BitTorrent Networks Performance Mechanisms , 2006, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[15]  Dan Wing,et al.  Session Traversal Utilities for NAT (STUN) , 2020, RFC.

[16]  Jonathan D. Rosenberg,et al.  Traversal Using Relays around NAT (TURN): Relay Extensions to Session Traversal Utilities for NAT (STUN) , 2020, RFC.

[17]  Konstantinos Psounis,et al.  Performance analysis of BitTorrent-like systems with heterogeneous users , 2007, Performance evaluation (Print).