ABSTRACT
Computational seismology is an area of wide sociological and economic impact, ranging from earthquake risk assessment to subsurface imaging and oil and gas exploration. At the core of these simulations is the modeling of wave propagation in a complex medium. Here we report on the extension of the high-order finite-element seismic wave simulation package SPECFEM3D to support the largest scale hybrid and homogeneous supercomputers. Starting from an existing highly tuned MPI code, we migrated to a CUDA version. In order to be of immediate impact to the science mission of computational seismologists, we had to port the entire production package, rather than just individual kernels. One of the challenges in parallelizing finite element codes is the potential for race conditions during the assembly phase. We therefore investigated different methods such as mesh coloring or atomic updates on the GPU. In order to achieve strong scaling, we needed to ensure good overlap of data motion at all levels, including internode and host-accelerator transfers. Finally we carefully tuned the GPU implementation. The new MPI/CUDA solver exhibits excellent scalability and achieves speedup on a node-to-node basis over the carefully tuned equivalent multi-core MPI solver. To demonstrate the performance of both the forward and adjoint functionality, we present two case studies run on the Cray XE6 CPU and Cray XK6 GPU architectures up to 896 nodes: (1) focusing on most commonly used forward simulations, we simulate seismic wave propagation generated by earthquakes in Turkey, and (2) testing the most complex seismic inversion type of the package, we use ambient seismic noise to image 3-D crust and mantle structure beneath western Europe.
- L. Bautista-Gomez, N. Maruyama, D. Komatitsch, S. Tsuboi, F. Cappello, S. Matsuoka, and T. Nakamura. FTI: high performance fault tolerance interface for hybrid systems. In Proceedings of the Supercomputing 2011 Conference, 2011. Google ScholarDigital Library
- L. Carrington, D. Komatitsch, M. Laurenzano, M. M. Tikir, D. Michéa, N. Le Goff, A. Snavely, and J. Tromp. High-frequency simulations of global seismic wave propagation using SPECFEM3D GLOBE on 62K processors. In Proceedings of the Supercomputing 2008 Conference, 2008. Google ScholarDigital Library
- C. Cecka, A. J. Lew, and E. Darve. Assembly of finite element methods on graphics processors. International journal for numerical methods in engineering, 85(5):640--669, 2011.Google Scholar
- C. Chevalier and F. Pellegrini. Pt-scotch: A tool for efficient parallel graph ordering. Parallel Computing, 34(68):318--331, 2008. Google ScholarDigital Library
- T. Cubit. Cubit 13.2 users manual. Sandia National Laboratories, Albuquerque, NM, 2012.Google Scholar
- J.-J. Droux. An algorithm to optimally color a mesh. Computer Methods in Applied Mechanics and Engineering, 104(2):249--260, 1993.Google ScholarCross Ref
- D. Kirk, W. H. Wen-mei, and W. Hwu. Programming massively parallel processors: a hands-on approach. Morgan Kaufmann, 2010. Google ScholarDigital Library
- D. Komatitsch. Fluid-solid coupling on a cluster of GPU graphics cards for seismic wave propagation. Comptes Rendus de l'Acadmie des Sciences Mécanique, 339:125--135, 2011.Google ScholarCross Ref
- D. Komatitsch, G. Erlebacher, D. Göddeke, and D. Michéa. High-order finite-element seismic wave propagation modeling with MPI on a large GPU cluster. Joural of Computational Physics, 229(20):7692--7714, 2010. Google ScholarDigital Library
- D. Komatitsch, G. Erlebacher, D. Göddeke, and D. Michéa. Modeling the propagation of elastic waves using spectral elements on a cluster of 192 GPUs. Computer Science Research and Development, 25(1-2):75--82, 2010.Google ScholarCross Ref
- D. Komatitsch, D. Michéa, and G. Erlebacher. Porting a high-order finite-element earthquake modeling application to NVIDIA graphics cards using CUDA. Journal of Parallel and Distributed Computing, 69(5):451--460, 2009. Google ScholarDigital Library
- D. Komatitsch and J. Tromp. Introduction to the spectral element method for three-dimensional seismic wave propagation. Geophysical Journal International, 139(3):806--822, December 1999.Google ScholarCross Ref
- D. Komatitsch and J. Tromp. Spectral-element simulations of global seismic wave propagation-I. Validation. Geophysical Journal International, 149(2):390--412, May 2002.Google ScholarCross Ref
- D. Komatitsch, S. Tsuboi, J. Chen, and J. Tromp. A 14.6 billion degrees of freedom, 5 teraflop, 2.5 terabyte earthquake simulation on the Earth Simulator. Proceedings of the ACM/IEEE Supercomputing SC'2003 conference, 2003. on CD-ROM. Google ScholarDigital Library
- M. Krivelevich. Coloring random graphs--an algorithmic perspective. In Proc. 2nd Colloquium on Mathematics and Computer Science, pages 175--195, 2002.Google ScholarCross Ref
- R. Madariaga. Dynamics of an expanding circular fault. Bulletin of the Seismological Society of America, 66(3):639--666, 1976.Google Scholar
- I. Molinari and A. Morelli. Epcrust: a reference crustal model for the european plate. Geophysical Journal International, 185(1):352--364, 2011.Google ScholarCross Ref
- T. Nissen-Meyer, A. Fournier, and F. A. Dahlen. A 2-D spectral-element method for computing spherical-earth seismograms--I. Moment-tensor source. Geophysical Journal International, 168:1067--1093, 2007b.Google ScholarCross Ref
- D. Peter, D. Komatitsch, Y. Luo, R. Martin, N. Le Goff, E. Casarotti, P. Le Loher, F. Magnoni, Q. Liu, C. Blitz, T. Nissen-Meyer, P. Basini, and J. Tromp. Forward and adjoint simulations of seismic wave propagation on fully unstructured hexahedral meshes. Geophysical Journal International, 186(2):721--739, August 2011.Google ScholarCross Ref
- N. Pulido, A. Ojeda, K. Atakan, and T. Kubo. Strong ground motion estimation in the sea of marmara region (turkey) based on a scenario earthquake. Tectonophysics, 391(1):357--374, 2004.Google ScholarCross Ref
- J. F. Schaefer, L. Boschi, T. W. Becker, and E. Kissling. Radial anisotropy in the European mantle: Tomographic studies explored in terms of mantle flow. Geophysical Research Letters, 38(23):L23304, December 2011.Google ScholarCross Ref
- N.M. Shapiro, M. Campillo, L. Stehly, and M. H. Ritzwoller. High-resolution surface-wave tomography from ambient seismic noise. Science, 307(5715):1615, 2005.Google ScholarCross Ref
- A. Tarantola. Inverse Problem Theory. Society for Industrial and Applied Mathematics, Philadelphia, 2005.Google Scholar
- J. Tromp, Y. Luo, S. Hanasoge, and D. Peter. Noise cross-correlation sensitivity kernels. Geophysical Journal International, 183(2):791--819, 2010.Google ScholarCross Ref
- J. Tromp, C. Tape, and Q. Liu. Seismic tomography, adjoint methods, time reversal and banana-doughnut kernels. Geophysical Journal International, 160:195--216, 2005.Google ScholarCross Ref
- H. M. Tufo and P. F. Fischer. Terascale spectral element algorithms and implementations. Proceedings of the ACM/IEEE Supercomputing SC'1999 conference, 1999. on CD-ROM. Google ScholarDigital Library
- H. Zhu, Y. Luo, T. Nissen-Meyer, C. Morency, and J. Tromp. Elastic imaging and time-lapse migration based upon adjoint methods. Geophysics, 74:WCA167--WCA177, 2009.Google Scholar
- Forward and adjoint simulations of seismic wave propagation on emerging large-scale GPU architectures
Recommendations
Forward and adjoint simulations of seismic wave propagation on emerging large-scale GPU architectures
SC '12: Proceedings of the 2012 International Conference for High Performance Computing, Networking, Storage and AnalysisComputational seismology is an area of wide sociological and economic impact, ranging from earthquake risk assessment to subsurface imaging and oil and gas exploration. At the core of these simulations is the modeling of wave propagation in a complex ...
Accelerating the discontinuous Galerkin method for seismic wave propagation simulations using the graphic processing unit (GPU)-single-GPU implementation
We have successfully ported an arbitrary high-order discontinuous Galerkin (ADER-DG) method for solving the three-dimensional elastic seismic wave equation on unstructured tetrahedral meshes to an Nvidia Tesla C2075 GPU using the Nvidia CUDA programming ...
Seismic wave propagation simulations on low-power and performance-centric manycores
We propose solutions for seismic wave propagation simulations on manycoresWe compare the performance and energy efficiency of our solutions to optimized solutions for multicores and GPUsMPPA-256 consumes at least 77% less energy than other processorsOur ...
Comments