A traffic-efficient message forwarding approach in unstructured P2P networks

Abstract

With more and more peer-to-peer (P2P) applications being utilized, the P2P traffic accounts for the majority in Internet, and thus leading to the network congestion problem. Reducing the redundant propagations of messages is an effective approach for solving such problem in unstructured P2P networks. In this paper, we first define a novel message structure which contains the information of message propagation path, and then three operations, including the inheritance, supplement and collection, on the message transmission paths are proposed, based on which a node could forward the message to the nodes who have not received the message yet purposefully by using the information of the past received messages and the characteristics of space and time of node activities, and thus eliminating the bandwidth consumption problem caused by the flooding-based message propagation approaches. The simulation results show that our strategy could effectively reduce the number of the redundant messages without lowering the message coverage ratio.

Structuring communities for sharing human digital memories in a social P2P network

“…We have assumed the digital memories of each person are stored on and shared through their own peer or personal social network [2]. Considering the importance of the underlying network structure can have a significant effect on the performance of the network [32,39,30,24,22], a Memory Thread-based Communities approach has been proposed to address this issue. An MTC reflects memory threads in a similar way to the method we use to structure peers in these memory thread-based Communities.…”