Publicado em: 11/12/2012

Proposta de Tese em Redes de Computadores

UNIVERSIDADE FEDERAL DO RIO GRANDE DO SUL
INSTITUTO DE INFORMÁTICA
PROGRAMA DE PÓS-GRADUAÇÃO EM COMPUTAÇÃO
———————————————-
DEFESA DE PROPOSTA DE TESE

Aluno: Flávio Roberto Santos
Orientador: Prof. Dr. Luciano Paschoal Gaspary
Coorientador: Prof. Dr. Antonio Marinho Pilla Barcellos

Título: Slowing Down to Speed Up: Delaying Users to Increase Quality of Experience in Content Distribution Systems

Linha de Pesquisa: Redes de Computadores

Data: 18/12/2012
Horário: 15h30min
Local: Sala 220 (conselhos). Prédio 43412 – Instituto de Informática

Banca Examinadora:
Prof. Dr. Francisco Vilar Brasileiro (UFCG) – Por videoconferência
Profa. Dra. Jussara Marques de Almeida (UFMG) – Por videoconferência
Prof. Dr. Lisandro Zambenedetti Granville (UFRGS)

Presidente da Banca: Prof. Dr. Luciano Paschoal Gaspary

Abstract:
The Internet has become a large platform where users can interact and share contents. These contents are typically personal files or third-party productions, which are published and downloaded by users based on their own interests. Considering the increasing demand for efficient content sharing, modern and robust content distribution systems (CDS) need to be deployed and maintained. In the context of this thesis, CDS are defined as systems used for sharing any kind of content on the Internet. Two categories of CDS are underscored as the most popular ones: file sharing and streaming systems.
Peer-to-peer (P2P) architectures have emerged as a potential solution to improve content dissemination in CDS. The adoption of P2P architectures, in the context of CDS, has attracted attention from the scientific community to some interesting challenges, namely network topology optimization, bootstrap mechanisms, and service discovery. One particular interesting challenge, in the context of this thesis, refers to mechanisms employed to match user’s interests and published contents. This is important to guarantee good quality of experience (QoE) to users. Imprecise descriptions are likely to happen due to different users’ opinion or malicious behavior. There has been substantial research on means to mitigate content pollution in CDS, leading to the proposal of several generic countermeasure mechanisms. They try to identify and isolate suspicious content after they are published. One limitation of these solutions is the time required to detect malicious activities may be long enough to allow the wide pollution dissemination. Another limitation is the fact that previous solutions attempt to polarize contents in either polluted or not. Therefore, they do not take into account the inherent subjectivity behind the evaluation of shared contents.
The main objective of this thesis is to provide users a good QoE — by acting proactively in the early stages of content distribution life cycle — and reduce the effect of malicious interferences. To achieve that, the research work presented in this thesis was carried out in three main steps. In the first iteration towards a solution to protect users from malicious behaviors, we proposed a novel strategy that operates conservatively to avoid wide dissemination of polluted content in early stages. In the second iteration, we advanced our previous solution to cope with the subjectivity regarding content descriptions. In the third and last iteration, we generalized the pollution problem to a collusion problem. The set of experiments carried out to evaluate our solution comprises both real tests and simulations. During the progress of this work, the overall solution was improved (through iterations) to become more robust and consider different users behavior, always focusing on reproducing real environments.

Keywords: content distribution systems, peer-to-peer systems, file sharing, streaming systems, tagging systems, delaying mechanisms, content pollution, collusion attacks