In this paper, the authors provide a study of mobile peer-to-peer (MP2P) networks via a simulation evaluation and comparison of random cache placement (RAND), global benefit-based cache placement (GBCP), localized benefit-based cache placement (LBCP), and cluster-based cache placement (CBCP) strategies. The paper also presents schemes to combine cache placement with dynamic cache update via global cache update (with a central server) and local cache update (assuming each mobile node has a cache list of data requests from itself and its nearby nodes). With these schemes, a comparison study shows a dramatic reduction in the average hop count (AHC) for random placement without update and random placement with update, and local placement without update and local placement with update.
The paper is easy to understand throughout. It includes sections on the following topics: simulations; update schemes and the relationship between update and placement; an introduction to MP2P networks; an introduction to the problem of achieving efficient data sharing; an introduction to the work of the paper; related work and perspective on the innovation of the work; the data placement problem; the data placement algorithms; and future work and conclusions. Two easy-to-understand suggestions for further work are also presented.
This work will be useful to those who are interested in applying this research to MP2P networks, as well as to those who are interested in further research of these networks.