An Agent-Based Approach for Modelling Peer to Peer Networks

A promising modelling and simulation tool is Agent-based Modelling (ABM) that has proved to be an effective and powerful tool across a wide range of fields. However, its exploitation within the Peer to Peer (P2P) paradigm has only recently attracted attention. This paper shows that the use of ABM tools such as NetLogo can provide many advantages over classical P2P simulation tools. This has been demonstrated through an initial case study involving implementation and evaluation of a well-known, powerful unstructured P2P protocol, named Gia. Preliminary simulation results show that Netlogo can provide the user with a wide range of comparative benefits, such as ease of implementation, visualization, and real-time analysis. Moreover, it can be used in educational applications for understanding the complexity of P2P networks. In addition, NetLogo is shown to have the potential to provide more insightful performance measures and serve as a future ` participatory' e-learning or educational tool for studying various aspects of the complex Gia model. Finally, we present some preliminary results for a NetLogobased implementation of a recently reported, modified Gnutella-like (M-Gia) protocol, which considers additional parameters to improve consistency in mobile P2P networks.