Mapping and evaluation of ordinary differential equations on heterogeneous architectures

Abstract

Several critical human activities depend on the weather forecasting. Some of them are transportation, health, work, safety, and agriculture. Such activities require computational solutions for weather forecasting through numerical models. These numerical models must be accurate and allow the computers to process them quickly. In this project we continue our master's research, where we intend to couple and evaluate new numerical methods for the solution of Ordinary Differential Equations (ODE) to BRAMS - Brazilian developments on the Regional Atmospheric Modelling System to a heterogeneous system composed of Xeon (Intel) processors coupled to a reprogrammable circuit (FPGA) via PCIe bus. According to the studies in the literature, the chemical equation from the mass continuity equation is the most computationally demanding part. This term computes several EDOs and the current solution is sequential and has a low precision. With this research, we intend to improve time execution through parallel numerical methods implemented in hardware and also improve the accuracy of the current model. (AU)