Mapping Geospatial Applications onto Parallel and Distributed Environments

The execution of Geospatial applications usually requires large computational and storage resources due to the massive amount of data, high resolutions, and large geographical areas they are using. Different parallel and distributed environments, such as Cluster, Multicore, Grid, and Cloud satisfy mostly the necessary requirements for running such applications. Depending on application features, data model, and processing requirements, one of such environments could be more appropriate and efficient, and could offer better performances than other ones. This paper presents a study and experiments on solutions for optimum mapping of the execution of Geospatial applications onto parallel and distributed environments. The research explores and highlights as well the elements by which such mapping solutions converge toward an optimum.