Redes BitTorrent (BT) atualmente representam o método Par-a-Par (P2P) de compartilhamento de arquivos pela Internet mais utilizado. Relatórios de monitoramento recentes revelam que as cópias de conteúdo sendo compartilhadas são, em grande maioria, ilegais e que filmes são os tipos de mídia mais populares. Iniciativas de pesquisa que tentaram entender a dinâmica da produção e do compartilhamento de conteúdo em redes BT não conseguiram prover informações precisas acerca da disseminação de cópias ilegais. No presente trabalho realizamos um extenso estudo experimental para caracterizar o comportamento de produtores, publicadores, provedores e consumidores de arquivos violando direitos autorais. O estudo conduzido é baseado em dados coletados durante sete meses de monitoração de enxames compartilhando filmes por meio de uma das comunidades públicas mais populares de BT. Os dados foram obtidos via emprego de uma arquitetura de monitoração do \universo" BitTorrent, o que permitiu popular uma base com informações acerca de mais de 55.000 torrents, 1.000 rastreadores e 1,9 milhões de IPs. Nossa análise não somente mostra que um pequeno grupo de usuários ativos _e responsável pela maior parte do compartilhamento de cópias ilegais, como desvenda relacionamentos existentes entre esses atores e caracteriza os padrões de consumo respeitados pelos usuários interessados nesse tipo de conteúdo. / BitTorrent (BT) networks are nowadays the most employed method of Peerto- Peer (P2P) le sharing in the Internet. Recent monitoring reports reveal that content copies being shared are mostly illegal and movies are the most popular media type. Research e orts carried out to understand the dynamics of content production and sharing in BT networks have been unable to provide precise information regarding the dissemination of illegal copies. In this work we perform an extensive experimental study in order to characterize the behavior of producers, publishers, providers and consumers of copyright-infringing les. This study is based on seven months of traces obtained by monitoring swarms sharing movies via one of the most popular BT public communities. Traces were obtained with an extension of a BitTorrent \universe" observation architecture, which allowed the collection of a database with information about more than 55,000 torrents, 1,000 trackers and 1.9 million IPs. Our analysis not only shows that a small group of active users is responsible for the majority of disseminated illegal copies, as it unravels existing relationships among these actors and characterizes consuming patterns respected by users interested in this particular set of contents.

Trabalhos recentes na literatura indicam que o BitTorrent é o protocolo de com- partilhamento de arquivos com maior popularidade, sendo responsável por mais da metade do tráfego P2P em algumas localidades geográficas. Apesar de vários estudos sobre a dinâmica do “universo BitTorrent”, até recentemente não existia metodologia para observá-lo sistematicamente. Um estudo preliminar indicou a existência de múltiplas estratégias de monito- ramento, que diferem em termos de objetos observados, conjunto de parâmetros e custos associados, e se sobrepõem em termos de informações extraídas. Um se- gundo trabalho apresentou uma combinação dessas estratégias na forma de uma arquitetura de monitoramento flexível e escalável. Nesse contexto, o objetivo da presente dissertação é investigar como otimizar o conjunto de estratégias e seus parâmetros para monitorar eficientemente o universo de redes BitTorrent tendo em vista um dado conjunto de informações a ser observado e recursos computacionais disponíveis. Como solução é proposto um controle de monitoramento adaptativo, que emprega um modelo de programação para otimizar, a cada rodada, o monitoramento considerando o estado percebido da rede. Embora o foco deste trabalho seja redes BitTorrent, é proposto um modelo genérico, que pode ser aplicado a outras redes P2P, aumentando portanto a contribuição desta dissertação. Os resultados de uma avaliação analítica indicam que o modelo de programação proposto gera soluções ótimas. Além disso, experimentos realizados com instâncias desse modelo geradas aleatoriamente mostram que o mesmo tem potencial para ser aplicado em redes mais complexas que BitTorrent. / Recent studies in the literature have indicated that BitTorrent is the most pop- ular file sharing protocol, being responsible for more than half of the P2P traffic in some geographical locations. Despite the several studies about the dynamics of the “BitTorrent universe”, until recently there has been no methodology to systematically observe it. A preliminary study identified multiple monitoring strategies, which differ in terms of observed objects, set of parameters and associated costs, and overlap in terms of extracted information. A subsequent work has combined those strategies in a flexible, extensible BitTorrent monitoring architecture. In this context, the goal of the present dissertation is to investigate how to optimize the set of strategies and their parameters for efficiently monitoring the universe of BitTorrent networks, considering a given set of monitoring objectives and available computational resources. As a solution, an adaptive monitoring control is proposed, which employs a programming model to optimize the monitoring at each round, considering the perceived network state. Although the focus of this work is BitTorrent networks, we propose a generic model that can be applied to other P2P networks, thereby increasing the contribution of this dissertation. The results of an analytical evaluation indicate that the proposed programming model generates optimal solutions. Furthermore, experiments carried out with ran- domly generated instances of this model show that it has potential to be applied to more complex networks than BitTorrent.

Trabalhos recentes na literatura indicam que o BitTorrent é o protocolo de com- partilhamento de arquivos com maior popularidade, sendo responsável por mais da metade do tráfego P2P em algumas localidades geográficas. Apesar de vários estudos sobre a dinâmica do “universo BitTorrent”, até recentemente não existia metodologia para observá-lo sistematicamente. Um estudo preliminar indicou a existência de múltiplas estratégias de monito- ramento, que diferem em termos de objetos observados, conjunto de parâmetros e custos associados, e se sobrepõem em termos de informações extraídas. Um se- gundo trabalho apresentou uma combinação dessas estratégias na forma de uma arquitetura de monitoramento flexível e escalável. Nesse contexto, o objetivo da presente dissertação é investigar como otimizar o conjunto de estratégias e seus parâmetros para monitorar eficientemente o universo de redes BitTorrent tendo em vista um dado conjunto de informações a ser observado e recursos computacionais disponíveis. Como solução é proposto um controle de monitoramento adaptativo, que emprega um modelo de programação para otimizar, a cada rodada, o monitoramento considerando o estado percebido da rede. Embora o foco deste trabalho seja redes BitTorrent, é proposto um modelo genérico, que pode ser aplicado a outras redes P2P, aumentando portanto a contribuição desta dissertação. Os resultados de uma avaliação analítica indicam que o modelo de programação proposto gera soluções ótimas. Além disso, experimentos realizados com instâncias desse modelo geradas aleatoriamente mostram que o mesmo tem potencial para ser aplicado em redes mais complexas que BitTorrent. / Recent studies in the literature have indicated that BitTorrent is the most pop- ular file sharing protocol, being responsible for more than half of the P2P traffic in some geographical locations. Despite the several studies about the dynamics of the “BitTorrent universe”, until recently there has been no methodology to systematically observe it. A preliminary study identified multiple monitoring strategies, which differ in terms of observed objects, set of parameters and associated costs, and overlap in terms of extracted information. A subsequent work has combined those strategies in a flexible, extensible BitTorrent monitoring architecture. In this context, the goal of the present dissertation is to investigate how to optimize the set of strategies and their parameters for efficiently monitoring the universe of BitTorrent networks, considering a given set of monitoring objectives and available computational resources. As a solution, an adaptive monitoring control is proposed, which employs a programming model to optimize the monitoring at each round, considering the perceived network state. Although the focus of this work is BitTorrent networks, we propose a generic model that can be applied to other P2P networks, thereby increasing the contribution of this dissertation. The results of an analytical evaluation indicate that the proposed programming model generates optimal solutions. Furthermore, experiments carried out with ran- domly generated instances of this model show that it has potential to be applied to more complex networks than BitTorrent.

Redes BitTorrent (BT) atualmente representam o método Par-a-Par (P2P) de compartilhamento de arquivos pela Internet mais utilizado. Relatórios de monitoramento recentes revelam que as cópias de conteúdo sendo compartilhadas são, em grande maioria, ilegais e que filmes são os tipos de mídia mais populares. Iniciativas de pesquisa que tentaram entender a dinâmica da produção e do compartilhamento de conteúdo em redes BT não conseguiram prover informações precisas acerca da disseminação de cópias ilegais. No presente trabalho realizamos um extenso estudo experimental para caracterizar o comportamento de produtores, publicadores, provedores e consumidores de arquivos violando direitos autorais. O estudo conduzido é baseado em dados coletados durante sete meses de monitoração de enxames compartilhando filmes por meio de uma das comunidades públicas mais populares de BT. Os dados foram obtidos via emprego de uma arquitetura de monitoração do \universo" BitTorrent, o que permitiu popular uma base com informações acerca de mais de 55.000 torrents, 1.000 rastreadores e 1,9 milhões de IPs. Nossa análise não somente mostra que um pequeno grupo de usuários ativos _e responsável pela maior parte do compartilhamento de cópias ilegais, como desvenda relacionamentos existentes entre esses atores e caracteriza os padrões de consumo respeitados pelos usuários interessados nesse tipo de conteúdo. / BitTorrent (BT) networks are nowadays the most employed method of Peerto- Peer (P2P) le sharing in the Internet. Recent monitoring reports reveal that content copies being shared are mostly illegal and movies are the most popular media type. Research e orts carried out to understand the dynamics of content production and sharing in BT networks have been unable to provide precise information regarding the dissemination of illegal copies. In this work we perform an extensive experimental study in order to characterize the behavior of producers, publishers, providers and consumers of copyright-infringing les. This study is based on seven months of traces obtained by monitoring swarms sharing movies via one of the most popular BT public communities. Traces were obtained with an extension of a BitTorrent \universe" observation architecture, which allowed the collection of a database with information about more than 55,000 torrents, 1,000 trackers and 1.9 million IPs. Our analysis not only shows that a small group of active users is responsible for the majority of disseminated illegal copies, as it unravels existing relationships among these actors and characterizes consuming patterns respected by users interested in this particular set of contents.

Context. Online piracy is widespread, controversial and poorly understood social phenomena that affects content creators, owners, and consumers. Online piracy, born from recent, rapid ITC changes, raises legal, ethical, and business challenges. Content owners, authors and content consumers should benefit from better understanding of online piracy. Improved, better adapted to marketplace and ITC changes content distribution models should benefit content owners and audiences.Objectives. Investigate online piracy effect on pirated product sales. Improve understanding of online piracy behaviors and process scale.Methods. This observational study investigated movie-sharing effect on U.S. box office. Movie sharing was observed over BitTorrent network, the most popular peer-to-peer file-sharing network. Relationship between piracy and sales was analyzed using linear regression model.Results. File sharing was found to have a slightly positive correlation with U.S. box office sales during first few weeks after film release, and no effect afterwards. Most of newly released movies are shared over BitTorrent network. File sharing is a global, massive phenomenon.Conclusions. I conclude that online movie file sharing has no negative correlation on U.S. box office. Slightly positive movie sharing correlation to box office sales could have occurred because sharing rather informs, than substitutes cinema going.

Existing VoD solutions often rely on unicast to distribute content, which leads to a higher load on the VoD server as more nodes become interested in the content. In such case, P2P is an alternative way of distributing content since it makes better use of available resources in the network. In this report, several P2P structures are evaluated from an operators point of view. We believe BitTorrent is the most adequate protocol for a P2P solution in IPTV environments. Two BitTorrent clients have been implemented on an IP-STB as proof of concept to find out whether P2P is suited for IPTV environments. Several tests were conducted to evaluate the performance of both clients and to see if they were able to reach a sufficient throughput on the IP-STB. Based upon the tests and the overall impressions, we are convinced that this particular P2P protocol is well suited for IPTV environments. Hopefully, a client developed from scratch for the IP-STB will offer even greater characteristics. Further, we have studied how to share recorded content among IP-STBs. Such a design would probably have many similarities to BitTorrent since a central node needs to keep track of content; the IP-STBs take care of the rest. The report also brings up whether BitTorrent is suitable for streaming. We believe that the necessary changes required to obtain such functionality will disrupt the strengths of BitTorrent. Some alternative solutions are presented where BitTorrent has been extended with additional modules, such as a server.

In this comparative study based on experimentation it is demonstrated that the two evaluated machine learning techniques, Bayesian networks and random forests, have similar predictive power in the domain of classifying torrents on BitTorrent file sharing networks. This work was performed in two steps. First, a literature analysis was performed to gain insight into how the two techniques work and what types of attacks exist against BitTorrent file sharing networks. After the literature analysis, an experiment was performed to evaluate the accuracy of the two techniques. The results show no significant advantage of using one algorithm over the other when only considering accuracy. However, ease of use lies in Random forests’ favour because the technique requires little pre-processing of the data and still generates accurate results with few false positives.

BitTorrent is an attractive peer-to-peer technology that attempts to reduce load on file sharers by allowing downloaders to share content between themselves. BitTorrent's current focus is to provide users with a fast download, which requires the file sharer to serve a disproportionate amount of the file. We present a modification to the BitTorrent seeding algorithm that reduces the load on BitTorrent file sharers. Essentially, if a block of a file is already available from a significant number of peers, the file sharer refuses to share that block, forcing peers to get it from each other. Using this modification, we show that there is a trade-off between the server's expended upload bandwidth and a longer peer download time. We also show some cases where we reduce the server's load as well as maintain a competitive peer download time by increasing the availability of rare blocks.

Peer-to-peer network applications has been a tremendous success among end users and has therefore received much attention in academia and industry, as have illegal public file sharing in media. However, private peer-to-peer file sharing between family, friends and co-workers have attracted little interest from the research community. Existing approaches also limit the users by not allowing for native interaction with userspace applications. In this paper we ex- -plore how private file sharing can be made safe, fast and scalable without constraining the users in this aspect. We demonstrate the concept of a private file sharing application utilizing a decentralized peer-to-peer network overlay by creating a prototype with extreme program- ming as methodology. To maximize the freedom of users the network is accessed through a virtual file-system interface. The prototype proves this to be a valid approach and we hope readers can use this paper as a platform for further developments in this area. / Fildelningsapplikationer som använder peer-to-peer teknik har varit en enorm framgång blandslutanvändare och har därmed erhållit mycket uppmärksamhet från akademi och indus- tri, liksom olaglig fildelning fått inom media. Däremot har inte privat fildelning mellan vän- ner, arbetskamrater och kollegor tilldelats samma uppmärksamhet från forskningssamfundet. Nuvarande tillämpningar begränsar användaren genom att inte tillåta naturlig interaktion med användarapplikationer. I denna uppsats utforskar vi hur privat fildelning kan göras snabb, skalbar och säker utan att begränsa användaren ur den aspekten. Vi demonstrerar ett koncept- för privat fildelning som nyttjar decentraliserad peer-to-peer arkitektur m.h.a en prototyp som tagits fram med extreme programming som metodologi. För att maximera användarnas frihet nyttjas ett virtuellt filsystem som gränssnitt. Prototypen visar att vår tillämpning fungerar i praktiken och vi hoppas att läsaren kan använda vårt arbete som en plattform för fortsatt utveckling inom detta område.

在目前的網際網路中，點對點(P2P)網路的應用扮演了一個重要的角色。由於在點對點網路架構下，檔案分享系統中檔案的數量以及來源大量增加，造成使用者必須花費更多的時間找尋想要的資源，因此搜尋功能就顯得格外地重要。BitTorrent(BT)作為一個點對點檔案分享系統，用戶量不斷增加，已逐漸成為網路頻寬的主要消耗者之一，但是其協定中沒有提供搜尋的功能，而且檔案分散於各個檔案發佈站之間，以及各發佈站之間無法有效搜尋彼此擁有的檔案資源，導致使用者查詢時效率不佳。因此本研究期望建立一套語意式搜尋機制來幫助使用者解決上述的問題。藉由使用語意網技術(Semantic Web)，針對BT檔案分享系統設計本體論，進行資源描述和建立簡單的分類，利用檔案的metadata來提供搜尋的功能以及彙整各檔案分佈站的檔案資源，做為上述議題的一個解決方案，讓使用者能夠更有效率地找到更完整的資源。 / In current World Wide Web, P2P network application plays an important role. Because the number and origin of files increase greatly in file sharing system under the architecture of P2P, causing users to spend more time searching for the resources they want. Therefore, the search function appears especially important. BitTorrent(BT), one kind of P2P file sharing system, has more and more users and becomes one of the biggest consumer of the network bandwidth. But it doesn't support any search function and shared files disperse between many web sites. Furthermore, these web sites can't exchange the shared files they own efficiently. These problems result in inefficient search performance. My research expects to propose a semantic search method to solve the problems mentioned above. By the means of Semantic Web technology, we design the ontology of BT file sharing system to describe the resources and establish simple taxonomy. In addition, using the metadata of files to provide BT for search function and collect the shared files between web sites. Let users find the shared files more efficiently and completely.