Motivations

The communication architectures of distributed applications are based today on the client-server paradigm. However, excellent examples (see Napster and Gnutella) have already highlighted how it does not meet all the requirements of some classes of distributed applications due to technological limitations inherent in the paradigm itself: for example, clients cannot communicate with each other but only nor can the server communicate with a client until the client has established a connection to the server on its own initiative. The peer-to-peer paradigm overcomes these technological limitations and, at a strategic level, enables two new key features of the applications:

  • exploit the processing capacity of the user terminals by moving some functions and part of the application logic from the center (the servers) to the periphery (the clients);
  • exploit the communication capacity between application logic on the terminals to collaborate and decrease the processing weight on the application servers.

Although the scientific community and standardization bodies are very active on these issues, the peer-to-peer paradigm has not yet been sufficiently explored, it is often difficult to apply, and, above all, there are still no development platforms and run environments. adequate time to support it neither for fixed network scenarios nor, even less, for networks and mobile terminals.
This project intends to oversee the state of the art and emerging technologies in this sector, analyze the technical and business problems inherent in the peer-to-peer paradigm and, in the specific case of J2ME mobile terminals and mobile phone network, provide a solution. prototype based on the Open Source JADE project and creating a trial of an application of interest to operators (TIM).
The possibility of keeping a user model on the terminal, updating it with automatic learning mechanisms, and using it for applications with logic on the client terminal, will also be the subject of specific project studies and realizations since, also, at a strategic level, it would allow the positioning of the operator with the role of value-added service of supply, profiling and user model management to all federated service providers.