Univ., Diss.--Köln, 2007. / Literaturverz. S. 257 - 264.

Hui Ying Kin. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (leaves 62-64). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.10 / Chapter 1.1 --- Motivation --- p.10 / Chapter 1.2 --- P2P small-world solution --- p.11 / Chapter 1.3 --- Balance of this thesis --- p.12 / Chapter 2 --- Background --- p.13 / Chapter 2.1 --- Small World phenomenon --- p.13 / Chapter 2.2 --- Internet Flash Crowds --- p.16 / Chapter 2.3 --- Dynamics in the small world network --- p.17 / Chapter 3 --- Small-world Overlay Protocol --- p.19 / Chapter 3.1 --- Overview --- p.23 / Chapter 3.2 --- Join Cluster Protocol (JCP): --- p.25 / Chapter 3.3 --- Leave Cluster Protocol (LCP): --- p.29 / Chapter 3.4 --- Object Lookup Protocol (OLP): --- p.31 / Chapter 3.5 --- Experimental Results Of Comparing with Other Structured P2P Networks --- p.33 / Chapter 3.5.1 --- Performance of object lookup: --- p.33 / Chapter 3.5.2 --- Effect of object lookup performance under different net- work sizes and number of long links: --- p.34 / Chapter 3.5.3 --- Comparison of Clustering Coefficient: --- p.35 / Chapter 4 --- Protocols for Handling Flash Crowd --- p.39 / Chapter 4.1 --- Static Flash Crowd --- p.40 / Chapter 4.2 --- Dynamic Flash Crowd --- p.44 / Chapter 4.3 --- Experimental Results for Replicating Popular Object --- p.45 / Chapter 4.3.1 --- Comparison between Chord and SWOP --- p.46 / Chapter 4.3.2 --- Comparison on Queue Size --- p.48 / Chapter 4.3.3 --- Variation on object request rate --- p.49 / Chapter 4.3.4 --- Variation on Number of Long Link Neighbors (k) --- p.50 / Chapter 4.4 --- Experiment Results for Examining the Effects on Traffic Loadings --- p.51 / Chapter 5 --- Performance Analysis --- p.53 / Chapter 5.1 --- Lookup complexity of SWOP --- p.53 / Chapter 5.2 --- Average time used for replicating item to all clusters in SWOP --- p.56 / Chapter 6 --- Related Work --- p.59 / Chapter 7 --- Conclusion --- p.61

Ma Tianbai. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (leaves 49-51). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Incentive P2P System Overview --- p.6 / Chapter 3 --- Resource Distribution Mechanism --- p.11 / Chapter 4 --- Resource Competition Game --- p.22 / Chapter 4.1 --- Theoretical Competition Game --- p.22 / Chapter 4.2 --- Practical Competition Game Protocol --- p.26 / Chapter 5 --- Generalized Mechanism and Game --- p.33 / Chapter 5.1 --- Generalized Mechanism with Incentive --- p.33 / Chapter 5.2 --- Generalized Mechanism with Utility --- p.35 / Chapter 6 --- Experiments --- p.38 / Chapter 7 --- Conclusion --- p.48

Lin Wing Kai. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 90-96). / Abstracts in English and Chinese. / Abstract/ 摘要 --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- """We are not alone""" --- p.1 / Chapter 1.2 --- Definition of P2P systems --- p.3 / Chapter 1.2.1 --- Terminologies --- p.4 / Chapter 1.2.2 --- Principles --- p.5 / Chapter 1.3 --- From sharing to replication --- p.7 / Chapter 1.3.1 --- Replication: why and how --- p.7 / Chapter 1.3.2 --- Advantages of P2P replication systems --- p.8 / Chapter 1.3.3 --- Typical replication approaches --- p.10 / Chapter 1.3.4 --- Difficulties in replication: resource allocation and replication strategy --- p.10 / Chapter 1.3.5 --- Why do peers cooperate? --- p.12 / Chapter 1.4 --- Contribution of this thesis --- p.13 / Chapter 1.4.1 --- Thesis organization --- p.13 / Chapter 2 --- Background Study --- p.15 / Chapter 2.1 --- Introduction --- p.15 / Chapter 2.2 --- Overview of P2P systems --- p.16 / Chapter 2.2.1 --- The original story --- p.16 / Chapter 2.2.2 --- Switching to decentralization --- p.16 / Chapter 2.2.3 --- Peer availability --- p.17 / Chapter 2.2.4 --- Other than file sharing --- p.18 / Chapter 2.3 --- Understanding replication --- p.20 / Chapter 2.3.1 --- File availability redefined --- p.20 / Chapter 2.3.2 --- Storage requirement analysis --- p.21 / Chapter 2.3.3 --- MTTF analysis --- p.22 / Chapter 2.3.4 --- Replica placement --- p.24 / Chapter 2.3.5 --- Other performance enhancement schemes --- p.27 / Chapter 2.4 --- Understanding cooperation --- p.28 / Chapter 2.5 --- Discussions --- p.30 / Chapter 3 --- Performance of erasure code replication --- p.32 / Chapter 3.1 --- Introduction --- p.32 / Chapter 3.2 --- Parameters definition --- p.33 / Chapter 3.2.1 --- File availability: whole file replication --- p.33 / Chapter 3.2.2 --- File availability: erasure code replication --- p.34 / Chapter 3.2.3 --- Properties of erasure code replication --- p.35 / Chapter 3.2.4 --- Effects of replication parameters --- p.36 / Chapter 3.2.5 --- Optimal value of b --- p.39 / Chapter 3.2.6 --- Analytical derivation --- p.40 / Chapter 3.3 --- Some practical considerations --- p.42 / Chapter 3.3.1 --- Cost of erasure code replication --- p.42 / Chapter 3.3.2 --- Sensitivity analysis --- p.44 / Chapter 3.4 --- Concluding remarks --- p.45 / Chapter 4 --- Distributed replication strategies --- p.48 / Chapter 4.1 --- Introduction --- p.48 / Chapter 4.2 --- The P2P replication system --- p.50 / Chapter 4.2.1 --- Erasure code replication --- p.50 / Chapter 4.2.2 --- Peers modelling --- p.51 / Chapter 4.2.3 --- Resource allocation problem --- p.52 / Chapter 4.2.4 --- Replication goal --- p.54 / Chapter 4.3 --- Decentralized adaptation --- p.56 / Chapter 4.3.1 --- Neighbour discovery and parameters exchange --- p.56 / Chapter 4.3.2 --- Storage resource estimation --- p.57 / Chapter 4.4 --- Heuristic strategies --- p.58 / Chapter 4.4.1 --- Random strategy --- p.58 / Chapter 4.4.2 --- Group partition strategy --- p.59 / Chapter 4.4.3 --- Highest available first (HAF) strategy --- p.61 / Chapter 4.5 --- Case studies --- p.65 / Chapter 4.5.1 --- Simulation results --- p.66 / Chapter 4.6 --- Concluding remarks --- p.69 / Chapter 5 --- Before cooperation: why do peers join? --- p.72 / Chapter 5.1 --- Introduction --- p.72 / Chapter 5.2 --- Information sharing club (ISC) model --- p.73 / Chapter 5.3 --- An example: music information sharing club --- p.75 / Chapter 5.4 --- Necessary condition for ISC to grow --- p.76 / Chapter 5.4.1 --- Music information sharing club example with simple requests --- p.78 / Chapter 5.5 --- Concluding remarks --- p.81 / Chapter 6 --- Conclusion --- p.83 / Chapter A --- Proof in this thesis --- p.86 / Bibliography --- p.90

With the increasing growth of communication networks, systems, and devices,technology has driven much research and development on a variety of communication protocols, applications, and smart devices. As a result, a variety of heterogeneous networks, de facto and standards have emerged. In parallel, users are also demanding to seamlessly use any device on any network infrastructure within this heterogeneous environment. To go beyond this problem, one of solutions is deploying service entities to be served in/on the network. The services can be composed of many independent service entities to redirect the communication flow combined with pipelined transformations. The service in the network can start from an application to intercept the user flow in a range of formats, originating in different access networks, and deliver them appropriately formatted for a particular end terminal based on the other end user preferences. To provide independent services to be served within the network, we have created a decentralized virtual market-place, which facilitates a place for services to be traded, discovered, and composed between providers and consumers. This approach of virtual market based on a decentralized system is leading to many advantages and challenges compared to other traditional network services. This thesis presents a novel approach towards using market management techniques to improve cooperation among traders in the community, while enhancing the community-oriented network service architecture. Without the centralized control, the decentralized virtual market-place requires the inclusion of techniques to provide better incentives. Given such incentives, rational traders will choose to behave co-operatively and contribute their resources to maximize the efficiency of the community. This will allow an application with dramatically improved utility, efficiency, and robustness and hence enable whole new domains of use. The viability of the decentralized virtual market-place is demonstrated using a prototype implementation and simulation system. The results have clearly shown that the decentralized virtual market-place can improve and overcome the major problems of most existing network service models.

<p>Realisering av streamingtjenester i P2P-nettverk er en utfordring fordi slike nettverk er preget av dynamikk og stadige endringer i forhold til nodenes tilgjengelighet. Hvis en node som leverer en mediestrøm blir utilgjengelig, vil det oppstå avbrudd i avspillingen hos mottakernoden. Utgangspunktet for denne hovedoppgaven er et ønske om å kunne gjøre streaming med tilfredsstillende tjenestekvalitet til en realitet i P2P-nettverk. Målet er å spesifisere en metode som bidrar til å forbedre den opplevde ressurstilgjengeligheten i slike nettverk.</p><p>P2P-systemer kjennetegnes generelt ved at de består av noder på randen av Internett, at samtlige noder bidrar med ressurser og kommuniserer direkte med hverandre og at hver node er en selvstyrende enhet. Det finnes to hovedkategorier P2P-systemer, “ekte” og “hybride”, samt en rekke applikasjonsområder. Streaming innebærer at innholdsleveranse gjøres i sanntid. Sanntidsleveranse krever at data leveres til mottaker i tide og at pakker leveres i rett rekkefølge. I P2P-nettverk, hvor omgivelsene preges av dynamikk og heterogenitet, er dette en stor utfordring. Flere teknikker kan benyttes for å håndtere slike omgivelser, men utfordringene knyttet til tilgjengelighet framstår som uløste og samtidig kritiske med tanke på oppnåelig tjenestekvalitet.</p><p>Vårt bidrag i retning av å forbedre den opplevde ressurstilgjengeligheten er en metode som blant annet benytter multinodestreaming som leveranseteknikk. Multinodestreaming innebærer å dele en strøm i flere delstrømmer og hente disse fra forskjellige noder. Delstrømmene kan da flettes ved mottak og spilles av som én strøm. Konsekvensen av at en node blir utilgjengelig vil kun bli kvalitetsdegradering av avspillingen, og ikke totalt avspillingsbrudd. Tilgjengeligheten kan forbedres ytterligere ved å utføre effektiv overtakelse av tapte delstrømmer. Ved å implementere metoden i en prototyp, er det mulig å teste dens funksjon og effekt. Testene viser at kvalitetsdegraderingen på avspillingen, som følge av at en node blir utilgjengelig, kun er midlertidig hvis en annen node kan overta leveransen. Reetablering av en delstrøm kan gjøres på gjennomsnittlig 4,7 sekunder. Enkeltnodestreaming ville til sammenligning medført fullt avspillingsbrudd. Resultatene sannsynliggjør derfor at multinodestreaming kan bidra til å forbedre den opplevde ressurstilgjengeligheten for streamingtjenester i P2P-nettverk.</p>

Thesis (M.S.)--Baylor University, 2005. / Includes bibliographical references (p. 78-81).

Resource location is a necessary operation for computer applications. In large scale peer-to-peer systems, random search is a scalable approach for locating dynamic resources. Current peer-to-peer systems can be partitioned into those which rely upon the Internet for message routing and those which utilize an overlay network. These two approaches result in different connectivity topologies. This thesis analyzes the effect of topological differences on the effectiveness of random search. After demonstrating the benefits of an overlay network, we propose a hybrid approach for resource location. Our proposed protocol provides deterministic searching capabilities which can help prevent request failures for sensitive applications.

Realisering av streamingtjenester i P2P-nettverk er en utfordring fordi slike nettverk er preget av dynamikk og stadige endringer i forhold til nodenes tilgjengelighet. Hvis en node som leverer en mediestrøm blir utilgjengelig, vil det oppstå avbrudd i avspillingen hos mottakernoden. Utgangspunktet for denne hovedoppgaven er et ønske om å kunne gjøre streaming med tilfredsstillende tjenestekvalitet til en realitet i P2P-nettverk. Målet er å spesifisere en metode som bidrar til å forbedre den opplevde ressurstilgjengeligheten i slike nettverk. P2P-systemer kjennetegnes generelt ved at de består av noder på randen av Internett, at samtlige noder bidrar med ressurser og kommuniserer direkte med hverandre og at hver node er en selvstyrende enhet. Det finnes to hovedkategorier P2P-systemer, “ekte” og “hybride”, samt en rekke applikasjonsområder. Streaming innebærer at innholdsleveranse gjøres i sanntid. Sanntidsleveranse krever at data leveres til mottaker i tide og at pakker leveres i rett rekkefølge. I P2P-nettverk, hvor omgivelsene preges av dynamikk og heterogenitet, er dette en stor utfordring. Flere teknikker kan benyttes for å håndtere slike omgivelser, men utfordringene knyttet til tilgjengelighet framstår som uløste og samtidig kritiske med tanke på oppnåelig tjenestekvalitet. Vårt bidrag i retning av å forbedre den opplevde ressurstilgjengeligheten er en metode som blant annet benytter multinodestreaming som leveranseteknikk. Multinodestreaming innebærer å dele en strøm i flere delstrømmer og hente disse fra forskjellige noder. Delstrømmene kan da flettes ved mottak og spilles av som én strøm. Konsekvensen av at en node blir utilgjengelig vil kun bli kvalitetsdegradering av avspillingen, og ikke totalt avspillingsbrudd. Tilgjengeligheten kan forbedres ytterligere ved å utføre effektiv overtakelse av tapte delstrømmer. Ved å implementere metoden i en prototyp, er det mulig å teste dens funksjon og effekt. Testene viser at kvalitetsdegraderingen på avspillingen, som følge av at en node blir utilgjengelig, kun er midlertidig hvis en annen node kan overta leveransen. Reetablering av en delstrøm kan gjøres på gjennomsnittlig 4,7 sekunder. Enkeltnodestreaming ville til sammenligning medført fullt avspillingsbrudd. Resultatene sannsynliggjør derfor at multinodestreaming kan bidra til å forbedre den opplevde ressurstilgjengeligheten for streamingtjenester i P2P-nettverk.

The objective of this project is to improve the performance of the Gnutella peer-to-peer protocol (version 0.4) by introducing a semantic-social routing model and several categories of interest. The Gnutella protocol requires peers to broadcast messages to their neighbours when they search files. The message passing generates a lot of traffic in the network, which degrades the quality of service. We propose using social networks to optimize the speed of search and to improve the quality of service in a Gnutella based peer-to-peer environment. Each peer creates and updates a friends list from its past experience, for each category of interest. Once peers generate their friends lists, they use these lists to semantically route queries in the network. Search messages in a given category are mainly sent to friends who have been useful in the past in finding files in the same category. This helps to reduce the search time and to decrease the network traffic by minimizing the number of messages circulating in the system as compared to standard Gnutella. This project will demonstrate by simulating a peer-to-peer type of environment with the JADE multi-agent system platform that by learning other peers interests, building and exploiting their social networks (friends lists) to route queries semantically, peers can get more relevant resources faster and with less traffic generated, i.e. that the performance of the Gnutella system can be improved.