Energy-efficiency and performance comparison of aerosol optical depth retrieval on distributed embedded SoC architectures

Abstract

The Aerosol Optical Depth (AOD) is a significant optical property of aerosols and is applied to the atmospheric correction of remotely sensed surface features as well as for monitoring volcanic eruptions, forest fires, and air quality in general, as well as gathering data for climate predictions on the basis of observations from satellites. We have developed an AOD retrieval workflow for processing satellite data not only with ordinary CPUs but also with parallel processors and GPU accelerators in a distributed hardware environment. This workflow includes pre-processing procedures which are followed by the runtime dominating main retrieval method. In this paper, we investigate if and how the main retrieval method can accommodate recent upcoming embedded hardware architectures in the field of high performance computing. We analyze and confirm the achieved performance as well as energy efficiency with real-world data from the moderate-resolution imaging spectro-radiometer (MODIS) and even compare the potential of those new architectures to today’s commonly available HPC hardware. Due to the very low energy intake, such embedded hardware architectures provide a great chance for situations with strong energy constraints like the pre-processing of recorded data on board of satellites.