STARR-DCS: Spatio-Temporal Adaptation of Random Replication for Data-Centric Storage

This article presents a novel framework for data-centric storage (DCS) in a wireless sensor and actor network (WSAN) that employs a randomly selected set of data replication nodes, which also change over time. This enables reductions in the average network traffic and energy consumption by adapting the number of replicas to applications' traffic, while balancing energy burdens by varying their locations. To that end, we propose and validate a simple model to determine the optimal number of replicas, in terms of minimizing average traffic/energy consumption, based on measurements of applications' production and consumption traffic. Simple mechanisms are proposed to decide when the current set of replication nodes should be changed, to enable new applications and nodes to efficiently bootstrap into a working WSAN, to recover from failing nodes, and to adapt to changing conditions. Extensive simulations demonstrate that our approach can extend a WSAN's lifetime by at least 60%, and up to a factor of 10× depending on the lifetime criterion being considered. The feasibility of the proposed framework has been validated in a prototype with 20 resource-constrained motes, and the results obtained via simulation for large WSANs have been also corroborated in that prototype.

STARR-DCS: Spatio-Temporal Adaptation of Random Replication for Data-Centric Storage Articles