A New Parallel 3D Front Propagation Algorithm for Fast Simulation of Geological folds

详细信息	查看全文 | 推荐本文 |

• 作者：Tor Gillberg^a ; ^c ; ^{torgi@simula.no} ; Mohammed Sourouri^a ; ^b ; ^{mohamso@ifi.uio.no} ; Xing Cai^a ; ^b ; ^{xingca@simula.no}
• 关键词：Static Hamilton-Jacobi equations ; front propagation ; automated C-to-CUDA code translation
• 刊名：Procedia Computer Science
• 出版年：2012
• 期刊代码：pc74_18770509
• 类别：cp
• 出版时间：2012
• 卷：9
• 期：Complete
• 页码：947-955
• 文件大小：501 K

摘要

We present a novel method for 3D anisotropic front propagation and apply it to the simulation of geological folding. The new iterative algorithm has a simple structure and abundant parallelism, and is easily adapted to multithreaded architectures using OpenMP. Moreover, we have used the automated C-to-CUDA source code translator, Mint, to achieve greatly enhanced computing speed on GPUs. Both OpenMP and CUDA implementations have been tested and benchmarked on several examples of 3D geological folding.