Математичні моделі та методи паралельних обчислень для розв’язання диференціальних рівнянь на графічних процесорах

Abstract

This work is focused on investigating the efficiency of parallel computing in modeling complex physical phenomena, such as fluid dynamics described by the Navier-Stokes equations. The main goal is to compare the performance of two implementations: based on a sequential (CPU) architecture, and on a parallel (GPU) one. The comparative analysis demonstrated that the GPU-based solver for incompressible inviscid fluid performs significantly better. These results highlight the advantages of GPUs in handling computationally intensive tasks and simulating fluid dynamics in real-time