A high performance remote sensing retrieval application on an institutional desktop Grid

Abstract

In this paper, we focus on making desktop Grids adapted to remote sensing application. The initial efforts were made to build a remote sensing retrieval application on the RSIN framework for dealing with climate change, named High Performance Aerosol property Retrieval Software (HiPARS). HiPARS uses integration of applications that access the backend of desktop Grid comprising of office PCs with a client. The HiPARS remote sensing application was built on RSIN framework, which uses the loose-coupled architecture of a desktop client and high throughput computing (HTC) and Grid backend. It allows for fast development by enabling existing code and new algorithms, and provides a familiar graphical environment for remote sensing users. The proposed solution is to transition the current PCs to a desktop Grid that can be cost-effectively sustained. It will be accessible for processing satellite data in quasi real-time, for parameter retrieving or for parallel processing efforts.