A Proximity-Aware Interest-Clustered P2P File Sharing System

Abstract

Efficient file question is vital to the general performance of peer-to-peer (P2P) file sharing systems.
Clustering peers by their common interests will considerably enhance the efficiency of file question. Clustering peers by
their physical proximity also can improve file question performance. However, few current works area unit able to
cluster peers supported each peer interest and physical proximity. Though structured P2Ps give higher file question
efficiency than unstructured P2Ps, it's tough to understand it as a result of their strictly outlined topologies. During this
work, we have a tendency to introduce a Proximity-Aware and Interest-clustered P2P file sharing System (PAIS)
supported a structured P2P, that forms physically-close nodes into a cluster and more teams physically-close and
common-interest nodes into a sub-cluster supported a graded topology. PAIS uses associate degree intelligent file
replication rule to more enhance file question efficiency. It creates replicas of files that area unit of times requested by a
gaggle of physically shut nodes in their location. Moreover, PAIS enhances the intra-sub-cluster file rummaging through
many approaches. First, it more classifies the interest of a sub-cluster to variety of sub-interests, and clusters commonsub interest nodes into a gaggle for file sharing. Second, PAIS builds associate degree overlay for every cluster that
connects lower capability nodes to higher capability nodes for distributed file querying whereas avoiding node overload.
Third, to cut back file looking out delay, PAIS uses proactive file info assortment in order that a file requester will
recognize if its requested file is in its close nodes. Fourth, to cut back the overhead of the file info assortment, PAIS uses
bloom filter primarily based file info assortment and corresponding distributed file looking out. Fifth, to boost the file
sharing efficiency, PAIS ranks the bloom filter leads to order. Sixth, considering that a recently visited file tends to be
visited once more, the bloom filter primarily based approach is increased by solely checking the freshly more bloom filter
info to cut back file looking out delay. Further, the experimental results show the high effectiveness of the intra-subcluster file looking out approaches in rising file looking out efficiency.