P2PGossip
Gossip-based algorithms, also called epidemic or probabilistic broadcast algorithms, have scalability and reliability properties that make them very appealing for disseminating information among a large number of nodes. They assure probabilistic guarantees and graceful degradation in the presence of failure. The underlying idea is very intuitive: every node that receives a message, buffers it, and then forwards it, that means gossiping it, a certain number of times, each time to a randomly selected subset of processes.
The main goal of the P2PGossip project is to evaluate gossip-based algorithms when they are applied to peer-to-peer systems, focusing on the dissemination of the information in order to determine the efficiency of this approach. This project is a performance evaluation study of gossip-based peer-to-peer systems.
P2PGossip is entirely developed in C for an efficiency purpose.
Gossip-based algorithms
Peer-to-peer
Epidemic algorithms
Peer sampling
Marco Graziosi