Achieving Energy Efficiency while Synchronizing Personal Data

As users increasingly rely on various portable de- vices and network connectivity to carry out com- puting, battery life will always be a major concern for users. This paper argues that energy-efficiency should be a first-class criteria for systems targeting personal environments. We examine how energy-efficiency can be achieved in data synchronization systems targeting this environment. We present general techniques that can be adopted by any synchronization system to save energy. In addition, we develop a novel cost model, based on weighted-energy-per-bit (Eb) and Eb forecasts, to help devices pick the right tech- nology, the right path and the right time to carry out synchronization. Our initial estimations indicate that up to two orders of magnitude of energy savings are possible depending on workloads.