DOI QR코드

DOI QR Code

Three-dimensional Wave Propagation Modeling using OpenACC and GPU

OpenACC와 GPU를 이용한 3차원 파동 전파 모델링

  • Kim, Ahreum (Department of Energy Resources Engineering, Pukyong National University) ;
  • Lee, Jongwoo (Department of Energy Resources Engineering, Pukyong National University) ;
  • Ha, Wansoo (Department of Energy Resources Engineering, Pukyong National University)
  • 김아름 (부경대학교 에너지자원공학과) ;
  • 이종우 (부경대학교 에너지자원공학과) ;
  • 하완수 (부경대학교 에너지자원공학과)
  • Received : 2017.02.20
  • Accepted : 2017.04.17
  • Published : 2017.05.31

Abstract

We calculated 3D frequency- and Laplace-domain wavefields using time-domain modeling and Fourier transform or Laplace transform. We adopted OpenACC and GPU for an efficient parallel calculation. The OpenACC makes it easy to use GPU accelerators by adding directives in conventional C, C++, and Fortran programming languages. Accordingly, one doesn't have to learn new GPGPU programming languages such as CUDA or OpenCL to use GPU. An OpenACC program allocates GPU memory, transfers data between the host CPU and GPU devices and performs GPU operations automatically or following user-defined directives. We compared performance of 3D wave propagation modeling programs using OpenACC and GPU to that using single-core CPU through numerical tests. Results using a homogeneous model and the SEG/EAGE salt model show that the OpenACC programs are approximately 53 and 30 times faster than those using single-core CPU.

3차원 주파수 영역과 라플라스 영역 파동장을 얻기 위해 시간 영역에서 파동 전파 모델링을 하는 동시에 푸리에 변환과 라플라스 변환을 수행하였다. 이 과정에서 효율적인 계산을 위해 OpenACC와 GPU를 이용한 병렬 연산을 수행하였다. OpenACC를 이용하면 기존의 C, C++, Fortran 등 프로그래밍 언어에 간단한 지시어(directive)를 추가하여 GPU 연산 가속기를 사용할 수 있기 때문에 CUDA 또는 OpenCL과 같은 GPGPU 프로그래밍 언어를 배우지 않고도 GPU를 이용한 프로그래밍을 할 수 있다. OpenACC 프로그램은 GPU 메모리 공간 할당, 호스트와 디바이스 간의 데이터 복사 및 GPU 연산 과정을 자동으로 또는 사용자 정의에 따라 수행하게 된다. 수치 실험으로 OpenACC와 GPU를 사용한 3차원 파동 전파 모델링 프로그램과 단일 CPU 코어를 사용한 프로그램의 성능을 비교하였다. 상속도 모델과 SEG/EAGE 암염돔 속도 모델을 이용한 결과, OpenACC와 GPU를 사용한 경우 단일 CPU 코어를 사용하였을 때보다 계산 속도가 각각 53배와 30배 정도 향상되었다.

Keywords

References

  1. Abdelkhalek, R., Calandra, H., Coulaud, O., Roman, J., and Latu, G., 2009, Fast seismic modeling and reverse time migration on a GPU cluster, International Conference on High Performance Computing & Simulation, 36-43.
  2. Choi, H., Seol, S., and Byun, J., 2015, Acceleration of anisotropic elastic reverse-time migration with GPUs, Geophys. and Geophys. Explor., 18, 74-84 (in Korean with English abstract). https://doi.org/10.7582/GGE.2015.18.2.074
  3. Furse, C. M., 2000, Faster than Fourier: ultra-efficient time-tofrequency-domain conversions for FDTD simulations, IEEE Antennas and Propagation, 42, 24-34. https://doi.org/10.1109/74.894179
  4. Kim, Y., Cho, Y., Jang, U., and Shin, C., 2013, Acceleration of stable TTI P-wave reverse-time migration with GPUs, Computers & Geosciences, 52, 204-217. https://doi.org/10.1016/j.cageo.2012.10.013
  5. Kim, A., Ryu, D., and Ha, W., 2016, Time-domain 3D wave propagation modeling and memory management using graphics processing units, Geophys. and Geophys. Explor., 19, 145-152 (in Korean with English abstract). https://doi.org/10.7582/GGE.2016.19.3.145
  6. Lee, H., Jeong, Y., and Lee, S, 2013, Survey on GPGPU programming models, Korea Information Processing Society, 20, 389-391.
  7. Lu, L., Renwei, D., Hongwei, L., and Hong, L., 2015, 3D hybrid-domain full waveform inversion on GPU, Computers & Geosciences, 83, 27-36. https://doi.org/10.1016/j.cageo.2015.06.017
  8. Mehra, R., Raghuvanshi, N., Savioja, L., Lin, M. C., and Manocha, D., 2012, An efficient GPU-based time domain solver for the acoustic wave equation, Applied Acoustics, 73, 83-94. https://doi.org/10.1016/j.apacoust.2011.05.012
  9. Nihei, K. T., and Li, X., 2007, Frequency response modelling of seismic waves using finite difference time domain with phase sensitive detection (TD-PSD), Geophysical Journal International, 169, 1069-1078. https://doi.org/10.1111/j.1365-246X.2006.03262.x
  10. OpenACC Programming and Best Practices Guide, 2015, http://www.openacc.org/content/openacc-programming-bestpractices-guide (April 14, 2017 Accessed)
  11. Ryu, D., Kim, A., and Ha, W., 2015, Expanding domain method for 3D time-Laplace-domain hybrid modeling, Geosystem Engineering, 18, 259-265. https://doi.org/10.1080/12269328.2015.1047966
  12. Shin, J., Ha, W., Jun, H., Min, D., and Shin, C., 2014, 3D Laplace-domain full waveform inversion using a single GPU card, Computers & Geosciences, 67, 1-13. https://doi.org/10.1016/j.cageo.2014.02.006
  13. Sirgue, L., Etgen, J., and Albertin, U., 2008, 3D frequency domain waveform inversion using time domain finite difference methods, 70th EAGE Conference and Exhibition Incorporating SPE EUROPEC.
  14. Wang, W., Xu, L., Cavazos, J., Huang, H. H., and Kay, M., 2014, Fast acceleration of 2D wave propagation simulations using modern computational accelerators, PLOS ONE, 9, e86484. https://doi.org/10.1371/journal.pone.0086484