Abstract
In this work, we examine the performance and energy efficiency when
using Python for developing HPC codes running on the GPU. We investigate the
portability of performance and energy efficiency between CUDA and OpenCL; beetween GPU generations; and between low-end, mid-range and high-end GPUs. Our
findings show that for some combinations of GPU and GPU code, there is a significant speedup for CUDA over OpenCL, but that this does not hold in general.
Our experiments show that performance in general varies more between different
GPUs, than between using CUDA and OpenCL. Finally, we show that tuning for
performance is a good way of tuning for energy efficiency.
using Python for developing HPC codes running on the GPU. We investigate the
portability of performance and energy efficiency between CUDA and OpenCL; beetween GPU generations; and between low-end, mid-range and high-end GPUs. Our
findings show that for some combinations of GPU and GPU code, there is a significant speedup for CUDA over OpenCL, but that this does not hold in general.
Our experiments show that performance in general varies more between different
GPUs, than between using CUDA and OpenCL. Finally, we show that tuning for
performance is a good way of tuning for energy efficiency.