Wong Wan Yeung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 118-122). / Abstracts in English and Chinese. / Chapter 1. --- Introduction --- p.1 / Chapter 1.1 --- Problem Definition --- p.1 / Chapter 1.2 --- Major Contributions --- p.5 / Chapter 1.2.1 --- S2S Searching --- p.6 / Chapter 1.2.2 --- GAroute --- p.8 / Chapter 1.3 --- Thesis Chapter Organization --- p.10 / Chapter 2. --- Related Work --- p.11 / Chapter 2.1 --- P2P Networks --- p.11 / Chapter 2.2 --- Query Routing Strategies --- p.20 / Chapter 2.3 --- P2P Network Security --- p.22 / Chapter 3. --- S2S Searching --- p.24 / Chapter 3.1 --- System Architecture --- p.24 / Chapter 3.1.1 --- Administration Module --- p.24 / Chapter 3.1.2 --- Search Module --- p.27 / Chapter 3.2 --- Indexing and Matching --- p.32 / Chapter 3.2.1 --- Background of Indexing and Matching --- p.32 / Chapter 3.2.2 --- Indexing Algorithm --- p.33 / Chapter 3.2.3 --- Matching Algorithm --- p.34 / Chapter 3.3 --- Query Routing --- p.36 / Chapter 3.3.1 --- Background of Query Routing --- p.36 / Chapter 3.3.2 --- Distributed Registrars and Content Summary --- p.38 / Chapter 3.3.3 --- Query Routing Algorithm --- p.41 / Chapter 3.3.4 --- Registrar Maintenance --- p.44 / Chapter 3.4 --- Communication Protocol --- p.45 / Chapter 3.4.1 --- Starting CGI --- p.46 / Chapter 3.4.2 --- Searching CGI --- p.47 / Chapter 3.4.3 --- Pinging CGI --- p.48 / Chapter 3.4.4 --- Joining CGI --- p.48 / Chapter 3.4.5 --- Leaving CGI --- p.48 / Chapter 3.4.6 --- Updating CGI --- p.49 / Chapter 3.5 --- Experiments and Discussions --- p.49 / Chapter 3.5.1 --- Performance of Indexing --- p.50 / Chapter 3.5.2 --- Performance of Matching --- p.52 / Chapter 3.5.3 --- Performance of S2S Searching --- p.54 / Chapter 3.5.4 --- Quality of Content Summary --- p.57 / Chapter 4. --- GAroute --- p.59 / Chapter 4.1 --- Proposed Hybrid P2P Network Model --- p.59 / Chapter 4.1.1 --- Background of Hybrid P2P Networks --- p.60 / Chapter 4.1.2 --- Roles of Zone Managers --- p.62 / Chapter 4.2 --- Proposed GAroute --- p.65 / Chapter 4.2.1 --- Genetic Representation --- p.69 / Chapter 4.2.2 --- Population Initialization --- p.70 / Chapter 4.2.3 --- Mutation --- p.72 / Chapter 4.2.4 --- Crossover --- p.74 / Chapter 4.2.5 --- Fission --- p.77 / Chapter 4.2.6 --- Creation --- p.80 / Chapter 4.2.7 --- Selection --- p.81 / Chapter 4.2.8 --- Stopping Criteria --- p.83 / Chapter 4.2.9 --- Optimization --- p.86 / Chapter 4.3 --- Experiments and Discussions --- p.89 / Chapter 4.3.1 --- Property of Different Topologies --- p.91 / Chapter 4.3.2 --- Scalability and Quality in Different Topologies --- p.92 / Chapter 4.3.3 --- Scalability and Quality in Different Quantities --- p.96 / Chapter 4.3.4 --- Verification of Lower Bandwidth Consumption --- p.101 / Chapter 4.3.5 --- Verification of Better Parallel Search --- p.105 / Chapter 5. --- Discussion --- p.110 / Chapter 6. --- Conclusion --- p.114 / Chapter 7. --- Bibliography --- p.118 / Chapter 8. --- Appendix --- p.123 / Chapter 8.1 --- S2S Search Engine --- p.123 / Chapter 8.1.1 --- Site Owner Perspective --- p.123 / Chapter 8.1.2 --- Search Engine User Perspective --- p.128 / Chapter 8.2 --- GAroute Library --- p.129

In this thesis, we attempt to revisit the problem of multi-party conferencing from a practical perspective, and tore think the design space involved in this problem. We believe that an emphasis on low end-to-end delays between any two parties in the conference is a must, and the source sending rate in a session should adapt to bandwidth availability and congestion. We present Celerity, a multi-party conferencing solution specifically designed to achieve our objectives. It is entirely Peer-to-Peer(P2P), and as such eliminating the cost of maintaining centrally administered servers. It is designed to deliver video with low end-to-end delays, at quality levels commensurate with available network resources over arbitrary network topologies where bottlenecks can be anywhere in the network. This is in contrast to commonly assumed P2P scenarios where bandwidth bottlenecks reside only at the edge of the network. The highlight in our design is a distributed and adaptive rate control protocol, that can discover and adapt to arbitrary topologies and network conditions quickly, converging to efficient link rate allocations allowed by the underlying network. In accordance with adaptive link rate control, source video encoding rates are also dynamically controlled to op-timize video quality in arbitrary and unpredictable network conditions. Celerity runs on the application layer and uses UDP to deliver the data. With the distributed rate control protocol, Celerity can deliver video at quality levels without the acknowledge of the underlying network topology, bandwidth, and the routing. We have implemented Celerity in a prototype system, and demonstrateits su¬perior performance over existing solutions in a local experimental test bed and over the Internet. In addition, using Celerity we have developed a multi-party conferencing system which provides real-time video and audio communication and allows users to dynamically join and leave, it achieves better user experience(low delay and high throughput) than existing products. / Chen, Xiangwen. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references (leaves 66-68). / Abstract --- p.i / Acknowledgement --- p.iii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation --- p.1 / Chapter 1.2 --- Contribution --- p.2 / Chapter 1.3 --- Thesis Organization --- p.4 / Chapter 2 --- Related Work --- p.5 / Chapter 3 --- Problem Formulation and Celerity Overview --- p.7 / Chapter 3.1 --- Settings --- p.7 / Chapter 3.2 --- Problem Formulation --- p.9 / Chapter 3.3 --- Celerity Overview --- p.10 / Chapter 4 --- Packing Delay-bounded Trees --- p.13 / Chapter 5 --- Overlay Link Rate Control --- p.18 / Chapter 5.1 --- Considering Both Delay and Loss --- p.18 / Chapter 5.2 --- A Loss-Delay Based Primal-Subgradient-Dual Algorithm --- p.20 / Chapter 5.3 --- Computing Subgradients of R{U+2098}(c{U+2098}, D) --- p.23 / Chapter 6 --- PRACTICAL IMPLEMENTATION --- p.26 / Chapter 6.1 --- Peer Functionality --- p.26 / Chapter 6.2 --- Critical Cut Calculation --- p.29 / Chapter 6.3 --- Utility Function --- p.29 / Chapter 6.4 --- Opportunistic Local Loss Recovery --- p.29 / Chapter 6.5 --- Fast Bootstrapping --- p.30 / Chapter 6.6 --- Operation Overhead --- p.30 / Chapter 6.7 --- Peer Computation Overhead --- p.31 / Chapter 6.8 --- System Implementation --- p.32 / Chapter 7 --- Experiments --- p.34 / Chapter 7.1 --- LAN Testbed Experiments --- p.34 / Chapter 7.1.1 --- Absence of Network Dynamics --- p.36 / Chapter 7.1.2 --- Cross Traffic --- p.39 / Chapter 7.1.3 --- Link Failure --- p.40 / Chapter 7.2 --- Peer Dynamics Experiments --- p.41 / Chapter 7.3 --- Internet Experiments --- p.42 / Chapter 8 --- Concluding Remarks --- p.46 / Chapter A --- Packing Delay-bounded Trees in the Presence of Helpers --- p.47 / Chapter B --- Proof of Theorem 1 and Theorem 3 --- p.50 / Chapter C --- Proof of Corollary 1 and Corollary 2 --- p.56 / Chapter D --- Proof of Proposition 1 --- p.58 / Chapter E --- Proof of Theorem 2 --- p.60 / Bibliography --- p.66

In peer-to-peer (P2P) systems, the phenomenon of churn (i.e., peer dynamics) will destroy the overlay structure, cause the loss of data objects, deteriorate the lookup performance, increase the bandwidth cost, and thus impact the performance of distributed applications greatly. Due to the prevalence of churn in real environments, it is essential to get a better understanding on how peer-to-peer systems evolve under churn and how to optimize the system performance under churn. In this thesis, we focus our research on the analysis and optimization of peer-to-peer systems under churn. Our research work falls into three main aspects: object storage under churn, object lookup under churn, and object (or load) balancing under churn. / Lastly, we study the effectiveness of two representative load balancing strategies in DHT-based P2P systems, (1) Rendezvous Directory Strategy (RDS) and (2) Independent Searching Strategy (ISS), under system churn. It enables us to have a clear understanding about their efficiency, scalability and robustness. Based on the analysis results, we also propose a Group Multicast Strategy (GMS) for load balancing in DHT systems, which attempts to achieve the benefits of both RDS and ISS. In order to have a better understanding of GMS, we also perform analytical studies on GMS in terms of its scalability and efficiency under churn. Finally, the effectiveness of GMS is evaluated by extensive simulation under different workload and churn levels. / Next, we consider the problem of optimizing lookup performance in DHT-based P2P systems under churn. We analytically study three important aspects on the optimization of DHT lookup performance, i.e., lookup strategy, lookup parallelism and lookup key replication. Our objective is to build a theoretical basis for the designers to make better choices in their future design. We first compare the performance of two representative lookup strategies - recursive routing and iterative routing, and explore the existence of better alternatives. Then we study the effectiveness of lookup parallelism in systems with different churn rates and show how to select the optimal degree of parallelism. Due to the importance of key replication on lookup performance, we also analyze the reliability of replicated keys under two different replication policies, and show how to perform proper configuration. Later, our results are also validated by simulation, and Kad is taken as a case to show the meaningfulness of our analysis. / We firstly develop a stochastic model to shed light on the evolution of stored objects in peer-to-peer systems under different types of churn, and analytically study the interplay between object maintenance and churn. To avoid the complexity of Markovian modeling, our model is based on stochastic differential equations, and thus we can provide closed-form terms to capture the system time-evolution, and formally derive asymptotic performance metrics of P2P storage systems under different maintenance strategies and various kinds of churn. Our analytical results provide some important insights in object maintenance under churn, which are useful in the optimization of P2P storage systems, e.g., reducing bandwidth usage, provisioning for bandwidth spike, improving system capacity, etc. Besides analytical study, our analysis is also validated by extensive simulation. / Wu, Di. / "July 2007." / Adviser: Kam-Wing Ng. / Source: Dissertation Abstracts International, Volume: 69-01, Section: B, page: 0443. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 174-188). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Zhao, Qiao. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 35-36). / Abstracts also in Chinese. / Abstract --- p.i / Acknowledgement --- p.v / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Model Description --- p.3 / Chapter 2.1 --- An Incentive Model for P2P Networks --- p.3 / Chapter 2.2 --- Learning Models for P2P Networks --- p.5 / Chapter 2.2.1 --- Current-best Learning Model (CBLM) --- p.5 / Chapter 2.2.2 --- Opportunistic Learning Model (OLM) --- p.6 / Chapter 2.3 --- Incentive Policies for P2P Networks --- p.7 / Chapter 2.3.1 --- Mirror Incentive Policy Vmirror --- p.8 / Chapter 2.3.2 --- Proportional Incentive Policy Vprop --- p.9 / Chapter 2.3.3 --- Linear Incentive Policy Class CLIP --- p.9 / Chapter 2.4 --- Performance and Robustness of Incentive Policies --- p.10 / Chapter 2.4.1 --- Robustness Analysis of Mirror Incentive Policy using the current-best learning method --- p.10 / Chapter 2.4.2 --- Robustness Analysis of Mirror Incentive Policy using the opportunistic learning method --- p.12 / Chapter 2.4.3 --- Robustness Analysis of Proportional Incentive Policy Using the current-best learning method --- p.12 / Chapter 2.4.4 --- Robustness Analysis of Proportional Incentive Policy Using the opportunistic learning method --- p.13 / Chapter 2.4.5 --- Robustness Analysis for Incentive Protocol in the Linear Incentive Class --- p.14 / Chapter 2.5 --- Connection with Evolutionary Game Theory --- p.17 / Chapter 3 --- Performance Evaluation --- p.21 / Chapter 3.1 --- Performance and Robustness of the Mirror Incentive Policy (Pmirror): --- p.21 / Chapter 3.2 --- Performance and Robustness of the Proportional Incentive Policy {Pprop): --- p.23 / Chapter 3.3 --- Performance and Robustness of incentive policy in the Linear Incentive Class (CLIP)： --- p.24 / Chapter 3.4 --- The Effect of Non-adaptive Peers: --- p.25 / Chapter 4 --- Adversary Effect of Altruism --- p.29 / Chapter 4.1 --- The Effect of Protocol Cost --- p.29 / Chapter 4.2 --- The Tradeoff between Altruism and System Robustness --- p.30 / Chapter 5 --- Related Work --- p.33 / Chapter 6 --- Conclusion --- p.34 / Bibliography --- p.35

Jia, Lu. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (p. 76-80). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Background Study --- p.7 / Chapter 3 --- Analysis of P2P Tracker Designs --- p.11 / Chapter 1 --- Tracker design in P2P systems --- p.11 / Chapter 1.1 --- A taxonomy of tracker designs --- p.11 / Chapter 1.2 --- Design considerations --- p.14 / Chapter 2 --- A reliability model for DHT-based tracker design --- p.15 / Chapter 2.1 --- DHT basics --- p.15 / Chapter 2.2 --- Model preliminaries and assumptions --- p.16 / Chapter 2.3 --- Model description --- p.18 / Chapter 3 --- Reliability analysis --- p.25 / Chapter 3.1 --- Related parameters --- p.25 / Chapter 3.2 --- Simulation setup --- p.27 / Chapter 3.3 --- Results --- p.30 / Chapter 3.4 --- Observations from modeling work --- p.35 / Chapter 3.5 --- Methods of DHT stabilization --- p.37 / Chapter 4 --- A Black-Box Study of Xunlei --- p.44 / Chapter 1 --- An Overview of Xunlei and its key components --- p.44 / Chapter 1.1 --- An overview --- p.44 / Chapter 1.2 --- Key components --- p.46 / Chapter 2 --- Participating into other swarms: Xunlei´ةs multi-protocol downloading strategy --- p.47 / Chapter 2.1 --- BitTorrent and eMule basics --- p.47 / Chapter 2.2 --- BitTorrent and eMule in Xunlei --- p.48 / Chapter 2.3 --- Multi-protocol downloading --- p.52 / Chapter 3 --- Xunlei servers --- p.54 / Chapter 4 --- Understanding Xunlei!s private protocol --- p.56 / Chapter 4.1 --- Exchanging peer lists --- p.56 / Chapter 4.2 --- Exchanging file data --- p.58 / Chapter 4.3 --- Error control and congestion control --- p.62 / Chapter 5 --- Further discussions --- p.65 / Chapter 5.1 --- Proximity of content --- p.65 / Chapter 5.2 --- Active swarm peers --- p.66 / Chapter 5.3 --- UDP-based data transmission --- p.69 / Chapter 5 --- Conclusion --- p.74 / Bibliography --- p.76

Fu, Zhengjia. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (p. 64-70 ). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Design for group network communication --- p.6 / Chapter 2.1 --- Performance metrics of network Voice Conference: GMOS --- p.7 / Chapter 2.2 --- Conference Leader Selection strategies --- p.11 / Chapter 2.3 --- Experiment Description --- p.14 / Chapter 2.4 --- Data analysis and results --- p.16 / Chapter 2.5 --- Applications of Proposals to Voice Conference --- p.25 / Chapter 3 --- P2P Application Identification --- p.27 / Chapter 3.1 --- Periodic Group Communication Patterns --- p.28 / Chapter 3.1.1 --- Terminology for Behavioral Patterns --- p.29 / Chapter 3.1.2 --- Pattern 1: Gossip of Buffer Maps --- p.30 / Chapter 3.1.3 --- Pattern 2: Content flow control --- p.31 / Chapter 3.1.4 --- Pattern 3: Synchronized Link Activation and Deactivation --- p.32 / Chapter 3.2 --- Identification Based on behavioral signatures --- p.33 / Chapter 3.2.1 --- Algorithm Overview --- p.34 / Chapter 3.2.2 --- Sequence Generation (SG1): Time Series for the Gossip Pattern --- p.36 / Chapter 3.2.3 --- Transform Time-domain Sequence to Frequency-domain Sequence --- p.36 / Chapter 3.2.4 --- Sequence Generation (SG2): Time Series for Content Flow Control Pattern --- p.40 / Chapter 3.2.5 --- Sequence Generation (SG3): Time Series for Synchronized Start and Finish of Flows --- p.41 / Chapter 3.2.6 --- Analyzer step --- p.47 / Chapter 3.3 --- Behavioral signatures of popular P2P applications --- p.47 / Chapter 3.4 --- Experiment Results --- p.49 / Chapter 3.5 --- Discussion --- p.52 / Chapter 4 --- Related Work --- p.58 / Chapter 5 --- Conclusion --- p.62 / Bibliography --- p.64

Li, Qiuhui. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 46-47). / Abstract also in Chinese. / Abstract --- p.i / Acknowledgement --- p.ii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- An overview of the BitTorrent protocol --- p.3 / Chapter 2 --- Problem Formulation --- p.7 / Chapter 2.1 --- Type-based Peer Selection Algorithm --- p.7 / Chapter 2.2 --- Clustering Index --- p.9 / Chapter 3 --- Model Formulation --- p.11 / Chapter 3.1 --- Markovian Model --- p.12 / Chapter 3.2 --- Transition Matrix --- p.14 / Chapter 3.2.1 --- Search Process --- p.16 / Chapter 3.2.2 --- Match Process --- p.18 / Chapter 3.2.3 --- Cut Process --- p.19 / Chapter 3.3 --- Open System Model --- p.21 / Chapter 4 --- Numerical Results and Observations --- p.24 / Chapter 4.1 --- Clustering Index --- p.24 / Chapter 4.2 --- Upload Utilization --- p.26 / Chapter 4.3 --- Download Rate --- p.28 / Chapter 4.4 --- Open System --- p.30 / Chapter 5 --- Performance Evaluation --- p.32 / Chapter 5.1 --- Model Verification --- p.34 / Chapter 5.2 --- Control the Clustering Index --- p.36 / Chapter 6 --- Related Works --- p.40 / Chapter 7 --- Conclusions --- p.44 / Bibliography --- p.46

Zhou, Yipeng. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 64-66). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Background --- p.1 / Chapter 1.2 --- Contribution --- p.2 / Chapter 1.3 --- Organization --- p.4 / Chapter 2 --- Related Work --- p.5 / Chapter 2.1 --- Work of Streaming --- p.5 / Chapter 2.2 --- Work of P2P VoD --- p.6 / Chapter 3 --- Basic Model of Synchronized Case --- p.8 / Chapter 4 --- Model of Chunk Selection Strategies --- p.13 / Chapter 4.1 --- Chunk Selection Strategies --- p.13 / Chapter 4.1.1 --- Greedy Strategy --- p.14 / Chapter 4.1.2 --- Rarest First Strategy --- p.15 / Chapter 4.1.3 --- "Buffer Size, Peer Population and Continuity" --- p.16 / Chapter 4.1.4 --- Mixed Strategy --- p.17 / Chapter 4.2 --- Some Conclusion and Extension --- p.19 / Chapter 4.3 --- Metrics --- p.20 / Chapter 4.3.1 --- Continuity --- p.20 / Chapter 4.3.2 --- Start-up Latency --- p.20 / Chapter 5 --- Experiment and Application --- p.22 / Chapter 5.1 --- Numerical Examples and Analysis --- p.22 / Chapter 5.2 --- Sensitivity study --- p.30 / Chapter 5.2.1 --- Discrete Model with Factor --- p.30 / Chapter 5.2.2 --- Validate Discrete Model with Factor --- p.31 / Chapter 5.2.3 --- Server Use Pull Strategy --- p.31 / Chapter 5.2.4 --- Vary Subset Size Touched by Server --- p.32 / Chapter 5.3 --- Application to Real-world Protocols --- p.32 / Chapter 6 --- Model of Unsynchronized Case --- p.34 / Chapter 6.1 --- The model for unsynchronized playback --- p.34 / Chapter 6.1.1 --- Overlap maximization problem --- p.37 / Chapter 6.1.2 --- Properties of the synchronized cluster --- p.38 / Chapter 6.2 --- Analysis of playback continuity --- p.40 / Chapter 6.2.1 --- Peers with different buffer sizes --- p.41 / Chapter 6.2.2 --- Analysis of two clusters with a lag --- p.44 / Chapter 7 --- Performance Evaluation of Unsynchronized System --- p.48 / Chapter 7.1 --- Performance Evaluation --- p.48 / Chapter 8 --- conclusion --- p.54 / Chapter 8.1 --- Conclusion --- p.54 / Chapter A --- Equation Derivation --- p.56 / Bibliography --- p.64

Ye, Chong. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves 61-63). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Classification of P2P systems --- p.1 / Chapter 1.2 --- Replication in P2P systems --- p.3 / Chapter 1.3 --- Related work --- p.7 / Chapter 1.4 --- Organization --- p.10 / Chapter 1.5 --- Publications derived from this work --- p.11 / Chapter 2 --- P2P Replication System --- p.12 / Chapter 2.1 --- Peers --- p.13 / Chapter 2.2 --- Erasure code replication --- p.13 / Chapter 2.3 --- Estimation of file availability --- p.14 / Chapter 2.4 --- Problem formulation --- p.15 / Chapter 2.5 --- Performance metrics --- p.17 / Chapter 3 --- Decentralized Decisions --- p.19 / Chapter 3.1 --- Writable peer set --- p.20 / Chapter 3.2 --- Stretch factor estimation --- p.21 / Chapter 3.3 --- Locking phase --- p.22 / Chapter 4 --- Equal Weight Replication --- p.23 / Chapter 4.1 --- The decentralized replication algorithms --- p.23 / Chapter 4.1.1 --- Random algorithm --- p.23 / Chapter 4.1.2 --- Group partition algorithm --- p.25 / Chapter 4.1.3 --- High Available First (HAF) algorithm --- p.27 / Chapter 4.2 --- Evaluation of algorithms --- p.29 / Chapter 4.2.1 --- Simulation setup --- p.29 / Chapter 4.2.2 --- Simulation results --- p.31 / Chapter 4.3 --- Summary --- p.40 / Chapter 4.4 --- Clarification of individual contributions --- p.41 / Chapter 5 --- Replication with Preferences --- p.43 / Chapter 5.1 --- Problem re-defined --- p.44 / Chapter 5.2 --- The Bi-weight model --- p.46 / Chapter 5.3 --- The statistical rounding policy --- p.48 / Chapter 5.4 --- The distributed replication algorithm --- p.49 / Chapter 5.5 --- Evaluation of the algorithm --- p.51 / Chapter 5.5.1 --- Simulation setup --- p.51 / Chapter 5.5.2 --- Simulation results --- p.52 / Chapter 5.6 --- Discussion --- p.56 / Chapter 5.7 --- Summary --- p.57 / Chapter 6 --- Future work --- p.58 / Chapter 7 --- Conclusion --- p.59 / Bibliography --- p.61

Cette thèse se base sur une vision originale des systèmes de Recherche d'Information en pair-à-pair (RI-P2P) ; pour proposer un ensemble d'outils pour mieux les évaluer et mieux les concevoir.Synthétiquement, la problématique d'évaluation consiste à comprendre quel système fonctionne bien dans quelles conditions ; comment les différents composants/aspects d'un système contribuent à ses performances ; comment des modifications du système peuvent faire évoluer ses performances. D'un point de vue conception, on voudrait savoir choisir une approche P2P en fonction du contexte de l'application à implémenter (jeu de données, propriétés du réseau...) ; comprendre comment mixer des systèmes ; mieux maîtriser l'impact d'une modification de l'implémentation sur le comportement et les performances d'un système.Nous pensons que ces deux problématiques d'évaluation et de conception sont fortement liées. Un même fondement est nécessaire pour progresser dans ces deux sens : mieux comprendre l'architecture des systèmes RI-P2P. À la base de notre réflexion, nous considérons qu'un système RI-P2P est l'implémentation en P2P d'un modèle RI, idée qui s'oppose au modèle traditionnel en deux couches. Nous nous appuyons sur cette idée pour donner un ensemble de définitions qui permettent de mieux appréhender les différents composants RI et composants P2P d'un même système et la façon dont ils s'organisent. Nous montrons ensuite comment on peut utiliser cette vision originale du domaine pour spécifier la démarche d'évaluation d'un système P2P et pour aider à la conception de nouveaux systèmes. / This PhD thesis develops an original vision of Peer-to-Peer Information Retrieval Systems. Based on this vision, we ropose a set of tools for a better evaluation and conception of P2P-IR systems.Synthetically, on the one hand, evaluating means understanding which system behaves well in which context ; how different components/aspects of a single system contribute to its performance ; how modifications of the system can make its performance evolve. On the other hand, considering design of P2P-IR systems, we would like to be able to choose a P2P approach based on what is known about the context the application will execute in (data set, network properties...) ; understand how to mix P2P systems ; better control the impact of a modification of the implementation of a system on its behavior an performance.We believe those concerns of evaluation and design are strongly related. There is a common need to improve on both directions : better understand the architecture of P2P-IR systems. On the basis of our reflexion, we consider a P2P-IR system is the implementation in P2P of an IR system, which contrasts with the traditional two-layered vision. We use this idea to give a set of definitions that help better understand the different RI and P2P components of a single system and their relationships. We then show how this vision can be used to help design new systems.