A case study of 3 D RTM-TTI algorithm on multicore and many-core platforms |
| |
Authors: | Zhang Xiuxia Tan Guangming Chen Mingyu Yao Erlin |
| |
Institution: | 1. State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, P.R.China;University of Chinese Academy of Sciences, Beijing 100049, P.R.China;2. State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, P.R.China |
| |
Abstract: | 3D reverse time migration in tiled transversly isotropic (3D RTM-TTI) is the most precise mod-el for complex seismic imaging.However, vast computing time of 3D RTM-TTI prevents it from be-ing widely used, which is addressed by providing parallel solutions for 3D RTM-TTI on multicores and many-cores.After data parallelism and memory optimization, the hot spot function of 3D RTM-TTI gains 35.99X speedup on two Intel Xeon CPUs, 89.75X speedup on one Intel Xeon Phi, 89.92X speedup on one NVIDIA K20 GPU compared with serial CPU baseline.This study makes RTM-TTI practical in industry.Since the computation pattern in RTM is stencil, the approaches also benefit a wide range of stencil-based applications. |
| |
Keywords: | 3D RTM-TTI Intel Xeon Phi NVIDIA K20 GPU stencil computing many-core multicore seismic imaging |
本文献已被 CNKI 万方数据 等数据库收录! |
|