Many applications today have the need for tracking changes and notifying users of changes when certain update thresholds are met. However, large-scale fresh information delivery systems have received very limited system software support. The proposed research aims at investigating research issues and evaluation methods for building decentralized Peer-to-Peer (P2P) information monitoring systems. The PI plans to develop a suite of system facilities at the Peer-to-Peer protocol layer to support large scale peer-to-peer information monitoring applications. Concretely, this research project will be carried out in two consecutive phases. First, it will undertake a systematic study of research issues in design and evaluation of a scalable peer-to-peer protocol and its supporting system architecture. Second, the project will conduct an in-depth investigation of technical and engineering issues in constructing an Internet-scale peer-to-peer information monitoring system PeerCQ. Specific issues of interest include load balance among heterogeneous peers, system utilization, mechanisms for efficient, scalable, and reliable processing of large numbers of information monitoring requests. The PeerCQ prototype will be delivered as an open source software to promote and enhance research and education in peer-to-peer computing and distributed information monitoring. The methodology and algorithms developed in this project will aid in the engineering, implementation, and evaluation of peer-to-peer information monitoring applications in real world.
