H. Astsatryan, A. Hayrapetyan, V. Narsisian, A. Saribekyan, Sh. Asmaryan, A. Saghatelyan, V. Muradyan, Y. Guigoz, G. Giuliani, N. Ray; Springer-Verlag "Earth Science Informatics", 2014, DOI: 10.1007/s12145-014-0165-3
The main objective of this paper is to introduce a portal of geoprocessing services that can be used to compute either a single vegetation index or a combination of vegetation indices, as a workflow. High Performance Computing (HPC) resources are used for the calculations, and the Web Processing Service (WPS) standard is used to handle the requests from and the responses to the portal. In case of a workflow, a single node of the cluster is dedicated to each index, and the number of used cores depends on the complexity of the task. In addition, based on a series of experiments made to accelerate remote sensing image processing, a parallelization method within the computational node is automatically chosen depending on the complexity of the operations and the amount of data. The suggested algorithm optimizes the processing by selecting the best methodology (serial or parallel) and the number of cores to efficiently manipulate and distribute the data. The interoperable web portal, Spatial Data Infrastructure (SDI) and the heterogeneous resources of HPC cluster are located in the same local area network, and the cluster nodes have access to the data via network file system sharing. The use of standardized web services makes it possible to use remote data as inputs.
Download PDF (1.1 KB)