Résumé

Many natural processes are characteristic multiscale multiphysics problems. Over the years techniques have been developed to study and simulate these processes using a computer. Such simulations are highly resource intensive and their performance is computationally bound on a large scale. High Performance Computing (HPC) plays a handy role in such a scenario. However, deploying a multiscale multiphysics application on a HPC infrastructure requires identifying and further tuning some parameters so as to improve performance and efficiency. This paper explains this procedure for the case study of a nutriment-driven coral growth process.

Détails

Actions

PDF