Uncertainty quantification through the Monte Carlo method in a cloud computing setting

摘要: The Monte Carlo (MC) method is the most common technique used for uncertainty quantification, due to its simplicity and good statistical results. However, computational cost extremely high, and, in many cases, prohibitive. Fortunately, MC algorithm easily parallelizable, which allows use simulations where computation of a single realization very costly. This work presents methodology parallelization method, context cloud computing. strategy based on MapReduce paradigm, an efficient distribution tasks cloud. illustrated problem structural dynamics that subject uncertainties. results show capable producing concerning moments low order. It shown even simple may require realizations convergence histograms, makes computing attractive (due high scalability capacity low-cost). Additionally, regarding time processing storage space usage allow one qualify this new as solution number beyond standard.