This paper describes a new approach for intensive rainfall data analysis. ITHACA’s Extreme Rainfall Detection System (ERDS) is conceived to provide near real-time alerts related to potential exceptional rainfalls worldwide, which can be used by WFP or other humanitarian assistance organizations to evaluate the event and understand the potentially floodable areas where their assistance is needed. This system is based on precipitation analysis and it uses rainfall data from satellite at worldwide extent. This project uses the Tropical Rainfall Measuring Mission Multisatellite Precipitation Analysis dataset, a NASA-delivered near real-time product for current rainfall condition monitoring over the world. Considering the great deal of data to process, this paper presents an architectural solution based on Grid Computing techniques. Our focus is on the advantages of using a distributed architecture in terms of performances for this specific purpose.