Abstract

 
 
 This paper presents an efficient parallel algorithm for the problem of converting satellite imagery in binary files. The algorithm was designed to update at global scale the land cover information used by the WRF climate model. We present the characteristics of the implemented algorithm, as well as the results of performance analysis and comparisons between two approaches to implement the algorithm. The performance analysis shows that the implemented parallel algorithm improves substantially against the sequential algorithm that solves the problem, obtaining a linear speedup.
 
 
Highlights
Wind prediction is crucial for many applications in environmental, energy, and economic contexts
We have designed and implemented two parallel implementations of the conversion algorithm, using shared memory and distributed memory approaches. Both parallel implementations of the conversion algorithm are currently operative in our cluster infrastructure (Cluster FING), Allowing to perform an efficient conversion of satellite images downloaded from NASA satellites, in order to update the information used by the Weather Research and Forecasting (WRF) climate model
This article presented an efficient algorithm for converting the full domain of land cover satellite images to the binary files within the WRF model
Summary
Wind prediction is crucial for many applications in environmental, energy, and economic contexts. In order to perform the soil information update for our country, but at planetary scale, a large number of satellite images need to be processed. In this context, applying high performance computing (HPC) techniques is a valuable strategy to reduce the execution time required to process the large volume of information in the images. The main contributions of the research reported in this article are: i) to introduce two parallel versions— using shared memory and distributed memory approaches—of an algorithm that process the satellite images to update the soil information used for wind prediction in the WRF model, and ii) to report an exhaustive experimental analysis that compares the computational efficiency of the shared and distributed memory parallel versions.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
