TY - GEN
T1 - Multi-GPU implementation of the horizontal diffusion method of the weather research and forecast model
AU - Solano-Quinde, Lizandro
AU - Gualan-Saavedra, Ronald
AU - Zuñiga-Prieto, Miguel
N1 - Publisher Copyright:
Copyright © 2016 ACM.
PY - 2016/3/12
Y1 - 2016/3/12
N2 - The Weather Research and Forecasting (WRF), a next generation mesoscale numerical weather prediction system, has a considerable amount of work regarding GPU acceleration. However, the amount of works exploiting multi-GPU sys- tems is limited. This work constitutes an effort on using GPU computing over the WRF model and is focused on a computationally intensive portion of the WRF: the Horizontal Diffusion method. Particularly, this work presents the enhancements that enable a single-GPU based implementation to exploit the parallelism of multi-GPU systems. The performance of the multi-GPU and single-GPU based implementations are compared on a computational domain of 433x308 horizontal grid points with 35 vertical levels, and the resulting speedup of the kernel is 3.5x relative to one GPU. The experiments were carried out on a multi-core computer with two NVIDIA Tesla K40m GPUs.
AB - The Weather Research and Forecasting (WRF), a next generation mesoscale numerical weather prediction system, has a considerable amount of work regarding GPU acceleration. However, the amount of works exploiting multi-GPU sys- tems is limited. This work constitutes an effort on using GPU computing over the WRF model and is focused on a computationally intensive portion of the WRF: the Horizontal Diffusion method. Particularly, this work presents the enhancements that enable a single-GPU based implementation to exploit the parallelism of multi-GPU systems. The performance of the multi-GPU and single-GPU based implementations are compared on a computational domain of 433x308 horizontal grid points with 35 vertical levels, and the resulting speedup of the kernel is 3.5x relative to one GPU. The experiments were carried out on a multi-core computer with two NVIDIA Tesla K40m GPUs.
KW - Compute Unified Device Archi- tecture (CUDA)
KW - Dynamic solver
KW - GPGPU
KW - Graphics Processing Unit (GPU)
KW - Horizon- tal diffusion method
KW - Multi-GPU implementation
KW - NVIDIA tesla K40m GPU
KW - Weather Research and Forecasting (WRF) model
UR - https://www.scopus.com/pages/publications/84968754875
U2 - 10.1145/2883404.2883407
DO - 10.1145/2883404.2883407
M3 - Contribución a la conferencia
AN - SCOPUS:84968754875
T3 - Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores, PMAM 2016
SP - 98
EP - 103
BT - Proceedings of the 7th International Workshop on Programming Models and Applications for Multicores and Manycores, PMAM 2016
A2 - Balaji, Pavan
A2 - Leung, Kai-Cheung
PB - Association for Computing Machinery
T2 - 7th International Workshop on Programming Models and Applications for Multicores and Manycores, PMAM 2016
Y2 - 12 March 2016
ER -