GA_Chord: An improvement to Chord algorithm based on group autonomy in structed P2P network

摘要

As a classical structed P2P model, Chord offers many salient features. The key issue of strengthening the system is how to deal with the influence from the frequent change of the nodes and promote the query efficiency. To advance the Chord performance, we propose a new model based on group autonomy which is called GA_Chord. A new table is added to differentiate all nodes into many neighbor groups. At the same time, system selects a more stable neighbor group to be leader nodes which manage to distribute resources. Each node only maintains the inside-group Finger Table and the Neighbor Table. Compared with the traditional Chord, GA_Chord greatly improves the query efficiency. Simultaneously, the model decreases the system cost caused by nodes' frequent change.