Leveraging the accelerated processing units for seismic imaging: A performance and power efficiency comparison against CPUs and GPUs

Oil and gas companies rely on high performance computing to process seismic imaging algorithms such as reverse time migration. Graphics processing units are used to accelerate reverse time migration, but these deployments suffer from limitations such as the lack of high graphics processing unit memory capacity, frequent CPU-GPU communications that may be bottlenecked by the PCI bus transfer rate, and high power consumptions. Recently, AMD has launched the Accelerated Processing Unit (APU): a processor that merges a CPU and a graphics processing unit on the same die featuring a unified CPU-GPU memory. In this paper, we explore how efficiently may the APU be applicable to reverse time migration. Using OpenCL (along with MPI and OpenMP), a CPU/APU/GPU comparative study is conducted on a single node for the 3D acoustic reverse time migration, and then extended on up to 16 nodes. We show the relevance of overlapping the I/O and MPI communications with the computations for the APU and graphics processing unit clusters, that performance results of APUs range between those of CPUs and those of graphics processing units, and that the APU power efficiency is greater than or equal to the graphics processing unit one.

[2]  Henri Calandra,et al.  Fast seismic modeling and reverse time migration on a graphics processing unit cluster , 2012, Concurr. Comput. Pract. Exp..

[3]  Jairo Panetta,et al.  Accelerating Kirchhoff Migration by CPU and GPU Cooperation , 2009, 2009 21st International Symposium on Computer Architecture and High Performance Computing.

[4]  Rached Abdelkhalek.  Accélération matérielle pour l'imagerie sismique : modélisation, migration et interprétation , 2013 .

[5]  D. Roweth,et al.  Leveraging the Cray Linux Environment Core Specialization Feature to Realize MPI Asynchronous Progress on Cray XE Systems , 2012 .

[6]  Bo Li,et al.  The issues of prestack reverse time migration and solutions with Graphic Processing Unit implementation , 2012 .

[7]  Vidar Slåtten,et al.  379 Performance Optimizations for TTI RTM on GPU based Hybrid Architectures , 2013 .

[8]  Henri Calandra,et al.  On the efficiency of the accelerated processing unit for scientific computing , 2016, SpringSim.

[9]  Jean-Pierre Berenger,et al.  A perfectly matched layer for the absorption of electromagnetic waves , 1994 .

[10]  Ligang Lu,et al.  Multi-level parallel computing of reverse time migration for seismic imaging on blue Gene/Q , 2013, PPoPP '13.

[11]  Gerhard Wellein,et al.  Prospects for truly asynchronous communication with pure MPI and hybrid MPI/OpenMP on current supercomputing platforms , 2011 .

[12]  Haohuan Fu,et al.  Selecting the right hardware for reverse time migration , 2010 .

[13]  Alexandre Denis,et al.  MPI Overlap: Benchmark and Analysis , 2016, 2016 45th International Conference on Parallel Processing (ICPP).

[14]  Asma Farjallah,et al.  Preparing depth imaging applications for Exascale challenges and impacts. (Etude de l'adéquation des machines Exascale pour les algorithmes implémentant la méthode du Reverse Time Migation) , 2014 .

[15]  J. Claerbout.  Toward a unified theory of reflector mapping , 1971 .

[16]  Issam Said,et al.  Contributions of hybrid architectures to depth imaging: a CPU, APU and GPU comparative study. (Apports des architectures hybrides à l'imagerie profondeur : étude comparative entre CPU, APU et GPU) , 2015 .

[17]  Paul L. Stoffa,et al.  3D Seismic Modeling And Reverse-Time Migration With the Parallel Fourier Method Using Non-blocking Collective Communications , 2009 .

[18]  R. Courant,et al.  On the Partial Difference Equations, of Mathematical Physics , 2015 .

[19]  Kirannmayi M. Sirasala,et al.  Experience of Porting and Optimization of Seismic Modelling on Multi and Many Cores of Hybrid Computing Cluster , 2015 .

[20]  T. Okamoto,et al.  Accelerating large-scale simulation of seismic wave propagation by multi-GPUs and three-dimensional domain decomposition , 2010 .

[21]  P. Moczo,et al.  The finite-difference time-domain method for modeling of seismic wave propagation , 2007 .

[22]  Dimitri Komatitsch,et al.  Accelerating a three-dimensional finite-difference wave propagation code using GPU graphics cards , 2010 .

[23]  Henri Calandra,et al.  Hybrid strategy for stencil computations on the APU , 2014 .

[24]  William W. Symes,et al.  Reverse time migration with optimal checkpointing , 2007 .

[25]  Peyman P. Moghaddam,et al.  Industrial-Scale Reverse Time Migration On GPU Hardware , 2009 .

[26]  Gerhard Wellein,et al.  Introduction to High Performance Computing for Scientists and Engineers , 2010, Chapman and Hall / CRC computational science series.

[27]  John C. Bancroft,et al.  Overcoming computational cost problems of reverse-time migration , 2010 .

[28]  Paulius Micikevicius,et al.  3D finite difference computation on GPUs using CUDA , 2009, GPGPU-2.

[29]  Hong Liu,et al.  The Algorithm of High Order Finite Difference Pre‐Stack Reverse Time Migration and GPU Implementation , 2010 .