Optimizing the satellite images execution process through a decision mediator approach

Earth Science applications are more and more present into our daily lives. Even if we talk about complex use cases (e.g. weather prediction, large geographic areas classification, etc.) or simple operations (e.g. relevant information extraction from a predefined data set) that are performed by regular users in different purposes, today's GIS application have to be scalable enough to process these requests in a reasonable amount of time. The current paper describes the GreenLand platform approach in solving these issues, by using a mediator module capable of redirecting the data execution to the Grid or standalone computing infrastructures.