high performance computing on graphics processing units: hgpu.org

Nowadays, a heat and mass transfer simulation plays an important role in various engineering and industrial fields. To analyze physical behaviors of a thermal environment, we have to simulate heat and mass transfer phenomena. However to obtain numerical solutions to heat and mass transfer equations is much time-consuming. In this paper, therefore, one of acceleration techniques developed in the graphics community that exploits a graphics processing unit (GPU) is applied to the numerical solutions of heat and mass transfer equations. Implementation of the simulation on GPU makes GPU computing power available for the most timeconsuming part of the simulation and calculation. The nVidia CUDA programming model provides a straightforward means of describing inherently parallel computations. This paper improves the computational performance of solving heat and mass transfer equations with the third kind boundary and initial conditions numerically running on GPU. We implemented simulation of heat transfer using the novel CUDA platform on nVidia Quadro FX 4800 and compared its performance with an optimized CPU implementation on a high-end Intel Xeon CPU. The experimental results of heat transfer clearly show that GPU can perform heat transfer simulation accurately and significantly accelerate the numerical calculation with the maximum observed speedups 20 times. Therefore, the GPU implementation is a promising approach to acceleration of the heat transfer simulation.