Title: Scalable Behavioral Simulation of Application-layer Peer-to-Peer Networking (Team 1)

The purpose of this project was to build a software implementation of Freenet to simulate the behaviour of a scalable application-layer peer-to-peer network. The project consisted of implementing the Freenet functionality into an existing network simulator. All milestones and objectives were completed successfully. Freenet is a type of decentralized peer-to-peer network which allows the distribution of uncensored information to its users. Anonymity of users is preserved by using Freenet’s data and message transferring protocols which does not allow data to be traced back to its originators. Freenet also uses routing algorithms to dynamically relocate data over the network from areas where the data is in low demand to areas where it is in high demand. These features can revolutionise the way the Internet is used and could lead to more efficient and faster networks. The project was implemented in C++ for the Linux operating system. The following features of Freenet were added to the existing simulator: message protocols, message and data routing algorithms, keys and searching algorithms, and management of data. Analysis of our test simulations conducted with 10, 100, 500 and 1000 node topologies prove that the implementation accurately follows the Freenet protocol. Time analysis of the results show that large and complex topologies can be simulated within a seconds. These test results support our claim that this implementation of Freenet on the skeleton is correct, fast and scalable