Experiments with low power commodity hardware platforms for challenged networks

Abstract

Wireless opportunistic connections are the primary mechanism for transferring data between disconnected nodes, such as vehicles, remote sensors, or village kiosks [6], [7] in a Delay Tolerant Network (DTN)[2]. Energy, being a scarce resource for such challenged environments, requires efficient management. We highlight the potential for accurate softwarelevel energy modeling and prediction on commodity embedded computers for DTN deployments, such as KioskNet [6]. Our experiments confirm that accurate prediction of energy consumption, without relying on any hardware power measurement tool, is feasible. We performed controlled micro-benchmarks on the three major components of an embedded computing device: the processor, the storage, and the communication sub-systems. Results show that energy consumption in all three follow simple models. Analysis of such measurements provide strong hints to the system designers about potential areas of optimization to make the device operations more power-aware.