An energy-efficient and client-centric data streaming technique on smartphones

Abstract

Today's users extensively download video files on their wireless handheld devices, which are inherently power constrained. Experiments have shown that video downloading accounts for a large portion of the total energy cost of downloading and playing video files. In this paper, we present a novel, energy-efficient and purely client-centric video downloading algorithm with three tunable parameters namely, buffer size, low water mark and socket-reading size. By implementing the algorithm on a smartphone and measuring the actual energy cost, we show the impacts of the three parameters on the energy cost of video downloads. By tuning the buffer size, low water mark, and socket-reading, we observed energy savings to the extent of 60%, 64%, and 63%, respectively. By incorporating our algorithmic approach into the process of video downloading, mobile application developers will be better positioned to fine-tune their apps to reduce the total energy cost of downloading large files in general.