Effects of Domain Partitioning on Parallel Computing of Atrium Airflow Simulation

Abstract

The accurate numerical simulation of large space airflow needs large computational quantity and long computation time usually. However, parallel computing is a good method to increase the speed of simulation. In this paper, the indoor airflow and thermal environment of a ventilated atrium with large space is simulated using the paralleling solver of FLUENT software, and the effect of two domain partitioning methods and various number of subdomains after domain partitioning on the simulated results are discussed. The calculated results show that the calculation time significantly reduced and the speedup rises from 1 to about 3 when the number of cores increases from 1 to 5. Furthermore, Principal Axes method has smaller wall times and higher speedups than Cartesian Y-Coordinate method. In addition, the number of subdomains should be controlled reasonably according to the number of cores because parallel computing needs the added boundary conditions through artificial hypothesis.