The direct energy demand of Internet data flows can be assessed using a variety of methodological approaches (top-down, bottom-up, or hybrid/model based) and different definitions of system boundaries. Because of this diversity, results reported in the literature differ by up to two orders of magnitude and are difficult to compare. We present a first assessment that uses a pure bottom-up approach and a system boundary that includes only transmission equipment. The assessment is based on the case study of a 40 megabit per second videoconferencing transmission between Switzerland and Japan, yielding a consumption of 0.2 kilowatt-hours per transmitted gigabyte for 2009, a result that supports the lowest of the existing estimates. We discuss the practical implications of our findings.

Experimental determination of both the short-time self-diffusion coefficient and the high-frequency viscosity are reported for model charge-stabilized dispersions. Despite the charged nature of the particles, we find that both quantities are insensitive to direct interparticle interactions. Quantitative agreement between the experimental data and micromechanical hard sphere theory is observed for a range of particle sizes and added salt content. However, while the reciprocal of the short-time self-diffusion coefficient shows a similar concentration dependence as the high-frequency viscosity, the two do not obey a generalized Stokes-Einstein relation. The violation is due to subtle differences in hydrodynamic interactions between the two particle motions, in agreement with theory and simulation.

No abstract

No abstract

No abstract