2MC-Match: A topology matching technique with 2-means clustering algorithm in P2P systems

A peer-to-peer (P2P) system is built upon an overlay network whose logical topology is independent of the underlying physical network. A lot of research work has been presented to address this issue, but most results still have some drawbacks, such as complexity or deployment difficulty. In order to alleviate the mismatching problem, we propose a topology matching technique with 2-means clustering algorithm called 2MC-Match, which uses the 2-means clustering algorithm to classify the Internet peers and build efficient ldquocloserdquo cluster. By performing the measured realistic Internet data (China), we show that 2MC-Match outperforms the well-known GNP in both aspects of accuracy and maintenance cost.