Bittorrent is one of the most popular and successful applications in the current Internet. However, we still have little knowledge about the topology of real Bittorrent swarms, how dynamic the topology is, and how it affects overall behavior. This paper describes an experimental study of the overlay topologies of real-world Bittorrent networks, focusing on the activity of the nodes of its P2P topology and especially their dynamic relationships. Peer Exchange Protocol (PEX) messages are analyzed to infer topologies and their properties, capturing the variations of their behavior. Our measurements, verified using the Kolmogorov-Smirnov goodness of fit test and the likelihood ratio test and confirmed via simulation, show that a power-law with exponential cutoff is a more plausible model than a pure power-law distribution. We also found that the average clustering coefficient is very low, supporting this observation. Bittorrent swarms are far more dynamic than has been recognized previously, potentially impacting attempts to optimize the performance of the system as well as the accuracy of simulations and analyses.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Mohamad Dikshie FAUZIE, Achmad Husni THAMRIN, Rodney VAN METER, Jun MURAI, "Assessing the Dynamics of Bittorrent Swarms Topologies Using the Peer Exchange Protocol" in IEICE TRANSACTIONS on Communications,
vol. E95-B, no. 5, pp. 1566-1574, May 2012, doi: 10.1587/transcom.E95.B.1566.
Abstract: Bittorrent is one of the most popular and successful applications in the current Internet. However, we still have little knowledge about the topology of real Bittorrent swarms, how dynamic the topology is, and how it affects overall behavior. This paper describes an experimental study of the overlay topologies of real-world Bittorrent networks, focusing on the activity of the nodes of its P2P topology and especially their dynamic relationships. Peer Exchange Protocol (PEX) messages are analyzed to infer topologies and their properties, capturing the variations of their behavior. Our measurements, verified using the Kolmogorov-Smirnov goodness of fit test and the likelihood ratio test and confirmed via simulation, show that a power-law with exponential cutoff is a more plausible model than a pure power-law distribution. We also found that the average clustering coefficient is very low, supporting this observation. Bittorrent swarms are far more dynamic than has been recognized previously, potentially impacting attempts to optimize the performance of the system as well as the accuracy of simulations and analyses.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E95.B.1566/_p
Copy
@ARTICLE{e95-b_5_1566,
author={Mohamad Dikshie FAUZIE, Achmad Husni THAMRIN, Rodney VAN METER, Jun MURAI, },
journal={IEICE TRANSACTIONS on Communications},
title={Assessing the Dynamics of Bittorrent Swarms Topologies Using the Peer Exchange Protocol},
year={2012},
volume={E95-B},
number={5},
pages={1566-1574},
abstract={Bittorrent is one of the most popular and successful applications in the current Internet. However, we still have little knowledge about the topology of real Bittorrent swarms, how dynamic the topology is, and how it affects overall behavior. This paper describes an experimental study of the overlay topologies of real-world Bittorrent networks, focusing on the activity of the nodes of its P2P topology and especially their dynamic relationships. Peer Exchange Protocol (PEX) messages are analyzed to infer topologies and their properties, capturing the variations of their behavior. Our measurements, verified using the Kolmogorov-Smirnov goodness of fit test and the likelihood ratio test and confirmed via simulation, show that a power-law with exponential cutoff is a more plausible model than a pure power-law distribution. We also found that the average clustering coefficient is very low, supporting this observation. Bittorrent swarms are far more dynamic than has been recognized previously, potentially impacting attempts to optimize the performance of the system as well as the accuracy of simulations and analyses.},
keywords={},
doi={10.1587/transcom.E95.B.1566},
ISSN={1745-1345},
month={May},}
Copy
TY - JOUR
TI - Assessing the Dynamics of Bittorrent Swarms Topologies Using the Peer Exchange Protocol
T2 - IEICE TRANSACTIONS on Communications
SP - 1566
EP - 1574
AU - Mohamad Dikshie FAUZIE
AU - Achmad Husni THAMRIN
AU - Rodney VAN METER
AU - Jun MURAI
PY - 2012
DO - 10.1587/transcom.E95.B.1566
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E95-B
IS - 5
JA - IEICE TRANSACTIONS on Communications
Y1 - May 2012
AB - Bittorrent is one of the most popular and successful applications in the current Internet. However, we still have little knowledge about the topology of real Bittorrent swarms, how dynamic the topology is, and how it affects overall behavior. This paper describes an experimental study of the overlay topologies of real-world Bittorrent networks, focusing on the activity of the nodes of its P2P topology and especially their dynamic relationships. Peer Exchange Protocol (PEX) messages are analyzed to infer topologies and their properties, capturing the variations of their behavior. Our measurements, verified using the Kolmogorov-Smirnov goodness of fit test and the likelihood ratio test and confirmed via simulation, show that a power-law with exponential cutoff is a more plausible model than a pure power-law distribution. We also found that the average clustering coefficient is very low, supporting this observation. Bittorrent swarms are far more dynamic than has been recognized previously, potentially impacting attempts to optimize the performance of the system as well as the accuracy of simulations and analyses.
ER -