No abstract

Abstrak Kata kunci: peer-to-peer jaringan, semantik kecil dunia, kecil dunia, gugus AbstractLocating content in unstructured peer-to-peer networks is a challenging problem. This paper presents a novel semantic small world resource search mechanism to address the problem. By using vector space model to compute the semantic relevance and applying small world properties such as low average hop distance and high clustering coefficient to construct a cluster overlay. In semantic small world system, the search mechanism is divided into two parts, searching at cluster and outside cluster through inner link and short link, so that it can achieve the incremental research. It significantly reduces the average path length and query cost. Meanwhile, the simulation results show that semantic small world scheme outperforms K-random walks and flooding scheme than higher query hit rate and lower query latency. Keywords: peer-to-peer network, semantic small world, semantic relevance, cluster IntroductionIn recent years, there have been many Peer-to-Peer (P2P) systems deployed in the Internet. The nodes which participate in sharing are increasing quickly and present a massive trend. In essence, a P2P system can be characterized as a distributed network system in which each node has similar functionalities and plays the role of a server and a client at the same time [1]. It is important for users to find relevance resources in the large scale P2P network. Content searching becomes a key step and a complicated problem in such environment.According to the nodes and resources organization and location method, P2P networks are divided into structured and unstructured. In a structured system, both the network topology and the distribution of keys are predetermined. They adopt DHT technology to indicate resource and queries, such as Chord [2] and Taperstry [3]. The advantage of such a structured system is high query efficiency. But DHT can't indicate the resources and query semantics, thus it can only be an exact match. On the other hand, unstructured systems like Gnutellla are fault-tolerant and resilient to the fuzzy query. But the search mechanism used in unstructured systems, like flooding, is consuming too much bandwidth and not scalable.Currently the unstructured peer-to-peer network is still the most widely used. Its basic search method is each node maintains a little neighbor information. Query messages are forwarded between nodes. Then match the query. According to whether there is prompt

AbstrakKeywords: chord protocol, P2P system, structure IntroductionWith the rapid development of network resource sharing, instant messaging, video transmission and other services, P2P networks have been widely used. The P2P network of distributed technology [1] has become a focus of international academic direction [2]. P2P network in accordance with the structure of the network can be divided into structured P2P network and unstructured P2P network, compared to unstructured P2P networks, structured P2P network no flooding resource search algorithm, which has a strong system scalability research value is also higher, including the chord network model is a structured P2P network in the most classic network model. This paper analysis the traditional chord model and some modifications chord model, then put forward a new improved chord model, though the theory and simulation analysis, prove the new chord model having a good efficiency.

No abstract