P2P File Diffusion and Interference (funded by arvato/Bertelsmann)

P2P File Diffusion and Interference

Research Project initiated and financed by

CMU Project Team

Heads

Phuoc Tran-Gia,

Kurt Tutschku

Researchers

Tobias Hossfeld,

Kenji Leibnitz (University of Osaka, Japan),

Rastin Pries,

Krzysztof Pawlikowski (University of Canterbury, New Zealand)

Summary

The volume of traffic data transmitted over the Internet has enormously increased recently due to the upcoming of peer-to-peer (P2P) file sharing applications. The most popular applications, such as Gnutella, eDonkey, or BitTorrent, are often abused for illegally sharing copyrighted content over the Internet. In P2P technology, each participant (peer) serves simultaneously as client and server which makes the system more scalable and robust and distinguishes it from conventional client-server architectures. However, this also comes at a slight drawback when considering content distribution. Since now, no longer a single trusted server distributes the file, malicious peers (interference peers) can offer fake or corrupted files and disrupt the file dissemination process.

There are two common approaches for the rightful owners of the files to protect their copyrighted property from being distributed. The first is to deliberately introduce files to the networks that are not indexed correctly. Usually, indexing is performed based on the file name, which in the case of such a poisoned file indicates a movie or song title other than the actual file. It is then mistakenly downloaded by other peers and the propagation of the intended file is slowed down. Another well known method is pollution. Here, deliberate corrupt versions of a file are injected to the network, which make use of the simple error correction methods of the file sharing applications. When the received data is recognized as corrupt, it is discarded and newly requested. This delays the overall downloading process and if the downloading delay exceeds the user's patience, he may become frustrated and abort the download.

We use non-stationary simulation techniques to predict the diffusion characteristics of single files shared in a P2P network. With the model we investigate the rate of diffusion, as well as the effects of pollution. We are particularly interested in the influence of these interference peers as they can greatly change the diffusion behaviour of the file. By modifying the population size of interference peers we can achieve approximately the same effects as performing a vaccination of the susceptible population. Additionally, our model takes the distinction between leechers and seeders into account and we show the influence of selfish peers on the file dissemination process. Especially, the ratio between seeders and interference peers and the willingness of the user to share the file are evaluated. The influence of parameters like sharing probability, request arrival rate, or file size on the diffusion process is also investigated.

Publications

On Pollution in eDonkey-like Peer-to-Peer File-Sharing Networks. Leibnitz, Kenji; Hoßfeld, Tobias; Wakamiya, Naoki; Murata, Masayuki. In 13th GI/ITG Conference on Measurement, Modeling, and Evaluation of Computer and Communication Systems, MMB2006. Nürnberg, Germany, 2006.
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
In this paper we propose an analytical model for file diffusion in a peer-to-peer (P2P) file-sharing network based on biological epidemics. During the downloading process, the peer shares the downloaded parts of the file and, thus, contributes to distributing it in the network. This behavior is similar to the spreading of epidemic diseases which is a well researched subject in mathematical biology. With our model we investigate the dynamics of file diffusion focusing on the effects of pollution, e.g. malicious peers sharing corrupted version of the file.

@inproceedings{info3-inproceedings-2006-207, abstract = {In this paper we propose an analytical model for file diffusion in a peer-to-peer (P2P) file-sharing network based on biological epidemics. During the downloading process, the peer shares the downloaded parts of the file and, thus, contributes to distributing it in the network. This behavior is similar to the spreading of epidemic diseases which is a well researched subject in mathematical biology. With our model we investigate the dynamics of file diffusion focusing on the effects of pollution, e.g. malicious peers sharing corrupted version of the file.}, address = {Nürnberg, Germany}, author = {Leibnitz, Kenji and Hoßfeld, Tobias and Wakamiya, Naoki and Murata, Masayuki}, booktitle = {13th GI/ITG Conference on Measurement, Modeling, and Evaluation of Computer and Communication Systems, MMB2006}, keywords = {p2pfdai}, month = {03}, title = {On Pollution in eDonkey-like Peer-to-Peer File-Sharing Networks}, year = 2006 }

%0 Conference Paper %1 info3-inproceedings-2006-207 %A Leibnitz, Kenji %A Hoßfeld, Tobias %A Wakamiya, Naoki %A Murata, Masayuki %B 13th GI/ITG Conference on Measurement, Modeling, and Evaluation of Computer and Communication Systems, MMB2006 %C Nürnberg, Germany %D 2006 %T On Pollution in eDonkey-like Peer-to-Peer File-Sharing Networks %X In this paper we propose an analytical model for file diffusion in a peer-to-peer (P2P) file-sharing network based on biological epidemics. During the downloading process, the peer shares the downloaded parts of the file and, thus, contributes to distributing it in the network. This behavior is similar to the spreading of epidemic diseases which is a well researched subject in mathematical biology. With our model we investigate the dynamics of file diffusion focusing on the effects of pollution, e.g. malicious peers sharing corrupted version of the file.
Modeling of Epidemic Diffusion in Peer-to-Peer File-Sharing Networks. Leibnitz, Kenji; Hoßfeld, Tobias; Wakamiya, Naoki; Murata, Masayuki. In The Second International Workshop on Biologically Inspired Approaches to Advanced Information Technology, Bio-ADIT 2006, LNCS 3853. Senri Life Science Center, Osaka, Japan, 2006.
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
In this paper we propose an analytical model for file diffusion in a peer-to-peer (P2P) file-sharing network based on biological epidemics. During the downloading process, the peer shares the downloaded parts of the file and, thus, contributes to distributing it in the network. This behavior is similar to the spreading of epidemic diseases which is a well researched subject in mathematical biology. Unlike other P2P models based on epidemics, we show that steady state assumptions are not sufficient and that the granularity of the diffusion model may be appropriately selected.

@inproceedings{info3-inproceedings-2006-204, abstract = {In this paper we propose an analytical model for file diffusion in a peer-to-peer (P2P) file-sharing network based on biological epidemics. During the downloading process, the peer shares the downloaded parts of the file and, thus, contributes to distributing it in the network. This behavior is similar to the spreading of epidemic diseases which is a well researched subject in mathematical biology. Unlike other P2P models based on epidemics, we show that steady state assumptions are not sufficient and that the granularity of the diffusion model may be appropriately selected.}, address = {Senri Life Science Center, Osaka, Japan}, author = {Leibnitz, Kenji and Hoßfeld, Tobias and Wakamiya, Naoki and Murata, Masayuki}, booktitle = {The Second International Workshop on Biologically Inspired Approaches to Advanced Information Technology, Bio-ADIT 2006, LNCS 3853}, keywords = {p2pfdai}, month = {01}, title = {Modeling of Epidemic Diffusion in Peer-to-Peer File-Sharing Networks}, year = 2006 }

%0 Conference Paper %1 info3-inproceedings-2006-204 %A Leibnitz, Kenji %A Hoßfeld, Tobias %A Wakamiya, Naoki %A Murata, Masayuki %B The Second International Workshop on Biologically Inspired Approaches to Advanced Information Technology, Bio-ADIT 2006, LNCS 3853 %C Senri Life Science Center, Osaka, Japan %D 2006 %T Modeling of Epidemic Diffusion in Peer-to-Peer File-Sharing Networks %X In this paper we propose an analytical model for file diffusion in a peer-to-peer (P2P) file-sharing network based on biological epidemics. During the downloading process, the peer shares the downloaded parts of the file and, thus, contributes to distributing it in the network. This behavior is similar to the spreading of epidemic diseases which is a well researched subject in mathematical biology. Unlike other P2P models based on epidemics, we show that steady state assumptions are not sufficient and that the granularity of the diffusion model may be appropriately selected.
Information Diffusion in eDonkey Filesharing Networks. Hoßfeld, Tobias. ATNAC 2004, Sydney, Australia, 2004, December 9.
- [ BibTeX ]
- [ EndNote ]
@presentation{info3-talk-2004-102, address = {ATNAC 2004, Sydney, Australia}, author = {Hoßfeld, Tobias}, keywords = {p2pfdai}, month = 12, title = {Information Diffusion in eDonkey Filesharing Networks}, year = 2004 }

%0 Generic %1 info3-talk-2004-102 %A Hoßfeld, Tobias %C ATNAC 2004, Sydney, Australia %D 2004 %T Information Diffusion in eDonkey Filesharing Networks
Information Diffusion in eDonkey Filesharing Networks. Hoßfeld, Tobias; Leibnitz, Kenji; Pries, Rastin; Tutschku, Kurt; Tran-Gia, Phuoc; Pawlikowski, Krzysztof. In ATNAC 2004, p. 8. Sydney, Australia, 2004.
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
Peer-to-Peer (P2P) applications have become highly popular in today’s Internet due to the spread of file sharing platforms such as KaZaa, eDonkey, and BitTorrent. In this paper we investigate the diffusion behavior of a file shared by multiple sources in an eDonkey-like peer-to-peer (P2P) environment. We provide two simulation models, one which captures the chunk transfer in a detailed way and a simplified model, where the upload bandwidth is equally split among all chunks of the file. The simulation of a single chunk transmission is sufficient to obtain accurate results, which leads to a much shorter simulation time. We further investigate the non-stationary process of file diffusion in a statistically reliable way. Depending on the popularity of the file, we consider flash crowd arrivals and constant arrivals. We derive simple analytical expressions showing the relationship between the propagation of the file, i.e. the number of peers sharing the chunk, and the main influencing factors: e.g. sharing probability, access probability, or arrival rate.

@inproceedings{info3-inproceedings-2004-150, abstract = {Peer-to-Peer (P2P) applications have become highly popular in today’s Internet due to the spread of file sharing platforms such as KaZaa, eDonkey, and BitTorrent. In this paper we investigate the diffusion behavior of a file shared by multiple sources in an eDonkey-like peer-to-peer (P2P) environment. We provide two simulation models, one which captures the chunk transfer in a detailed way and a simplified model, where the upload bandwidth is equally split among all chunks of the file. The simulation of a single chunk transmission is sufficient to obtain accurate results, which leads to a much shorter simulation time. We further investigate the non-stationary process of file diffusion in a statistically reliable way. Depending on the popularity of the file, we consider flash crowd arrivals and constant arrivals. We derive simple analytical expressions showing the relationship between the propagation of the file, i.e. the number of peers sharing the chunk, and the main influencing factors: e.g. sharing probability, access probability, or arrival rate.}, address = {Sydney, Australia}, author = {Hoßfeld, Tobias and Leibnitz, Kenji and Pries, Rastin and Tutschku, Kurt and Tran-Gia, Phuoc and Pawlikowski, Krzysztof}, booktitle = {ATNAC 2004}, keywords = {p2pfdai}, month = 12, pages = 8, title = {Information Diffusion in eDonkey Filesharing Networks}, year = 2004 }

%0 Conference Paper %1 info3-inproceedings-2004-150 %A Hoßfeld, Tobias %A Leibnitz, Kenji %A Pries, Rastin %A Tutschku, Kurt %A Tran-Gia, Phuoc %A Pawlikowski, Krzysztof %B ATNAC 2004 %C Sydney, Australia %D 2004 %P 8 %T Information Diffusion in eDonkey Filesharing Networks %X Peer-to-Peer (P2P) applications have become highly popular in today’s Internet due to the spread of file sharing platforms such as KaZaa, eDonkey, and BitTorrent. In this paper we investigate the diffusion behavior of a file shared by multiple sources in an eDonkey-like peer-to-peer (P2P) environment. We provide two simulation models, one which captures the chunk transfer in a detailed way and a simplified model, where the upload bandwidth is equally split among all chunks of the file. The simulation of a single chunk transmission is sufficient to obtain accurate results, which leads to a much shorter simulation time. We further investigate the non-stationary process of file diffusion in a statistically reliable way. Depending on the popularity of the file, we consider flash crowd arrivals and constant arrivals. We derive simple analytical expressions showing the relationship between the propagation of the file, i.e. the number of peers sharing the chunk, and the main influencing factors: e.g. sharing probability, access probability, or arrival rate.
Information Diffusion in eDonkey Filesharing Networks. Technical Report (341), . Hoßfeld, Tobias; Leibnitz, Kenji; Pries, Rastin; Tutschku, Kurt; Tran-Gia, Phuoc; Pawlikowski, Krzysztof. University of Wuerzburg, 2004.
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
@techreport{info3-techreport-2004-163, author = {Hoßfeld, Tobias and Leibnitz, Kenji and Pries, Rastin and Tutschku, Kurt and Tran-Gia, Phuoc and Pawlikowski, Krzysztof}, institution = {University of Wuerzburg}, keywords = {sys:hidden:techreport}, month = {09}, number = 341, title = {Information Diffusion in eDonkey Filesharing Networks}, year = 2004 }

%0 Report %1 info3-techreport-2004-163 %A Hoßfeld, Tobias %A Leibnitz, Kenji %A Pries, Rastin %A Tutschku, Kurt %A Tran-Gia, Phuoc %A Pawlikowski, Krzysztof %D 2004 %N 341 %T Information Diffusion in eDonkey Filesharing Networks

P2P File Diffusion and Interference (funded by arvato/Bertelsmann)