Neighborhood-Based Scalable Replica Allocation Scheme in Mobile Ad-hoc Networks

In mobile ad-hoc networks, data accessibility is lower than that in traditional fixed networks due to frequent network divisions. Several schemes have been proposed to improve the data accessibility by replicating data items. These schemes utilize grouping to achieve high data accessibility. Although the grouping is an effective way to obtain high data accessibility, it may lead to huge amount of communication cost. In this paper, we propose a scalable replica allocation scheme to reduce the communication cost while maintaining high data accessibility. The proposed scheme builds groups of nodes in the network in such a way that each group has the star network topology. The central node in each group plays a role of a replica allocator who distributes replicas to the group nodes based on the estimated data access frequencies that are computed periodically. We evaluate the proposed scheme by simulations using the network simulator NS-3 as well as analytical evaluations. The results of simulations and analytical evaluations demonstrate that our proposed scheme reduces the communication cost significantly over the existing replica allocation schemes as the number of users increases in the network, while achieving comparable data accessibility.