Parallel processing of Prestack Kirchhoff Time Migration on a PC Cluster

Dai, Hengchang

doi:10.1016/j.cageo.2005.02.002

Cited by 12 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The processing time is longer than the communication time, so the time elapsed is inversely proportional to the number of CPUs, and thus using more CPU nodes can reduce the time elapsed and improve the efficiency (Dai 2005). In this study, we propose a complete GPU solution for PSTM.…”

Section: Output: Common Reflection Gather I[ix][iy][ih][it]mentioning

confidence: 93%

“…However, the practical application of PSTM to tasks during large 3D surveys is still computationally intensive. To accelerate the processing of migration, the parallel processing of prestack time migration has been implemented routinely on distributed parallel computers (Schleicher andCopeland 1993, Chen et al 1993), as well as on PC clusters (Morton et al 1999, Hellman 2000, Dai 2005. In recent years, many other devices have also been used to accelerate PSTM such as FPGAs (He et al 2005).…”

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Practical Implementation of Prestack Kirchhoff Time Migration on a General Purpose Graphics Processing Unit

2016

Acta Geophys.

View full text Add to dashboard Cite

A b s t r a c tIn this study, we present a practical implementation of prestack Kirchhoff time migration (PSTM) on a general purpose graphic processing unit. First, we consider the three main optimizations of the PSTM GPU code, i.e., designing a configuration based on a reasonable execution, using the texture memory for velocity interpolation, and the application of an intrinsic function in device code. This approach can achieve a speedup of nearly 45 times on a NVIDIA GTX 680 GPU compared with CPU code when a larger imaging space is used, where the PSTM output is a common reflection point that is gathered as I[nx][ny] [nh][nt] in matrix format. However, this method requires more memory space so the limited imaging space cannot fully exploit the GPU sources. To overcome this problem, we designed a PSTM scheme with multi-GPUs for imaging different seismic data on different GPUs using an offset value. This process can achieve the peak speedup of GPU PSTM code and it greatly increases the efficiency of the calculations, but without changing the imaging result.

show abstract

Section: Output: Common Reflection Gather I[ix][iy][ih][it]mentioning

confidence: 93%

Section: Introductionmentioning

confidence: 98%

Practical Implementation of Prestack Kirchhoff Time Migration on a General Purpose Graphics Processing Unit

2016

Acta Geophys.

View full text Add to dashboard Cite

show abstract

“…Dai et al , introduced an embarrassingly parallel PKTM program to run in a single‐core cluster where each output trace is individually processed by each working node. Yerneni et al , studied computational demands of PKTM and presented a parallel implementation for a cluster of workstations.…”

Section: Related Workmentioning

confidence: 99%

Accelerating Pre‐stack Kirchhoff Time Migration by Manual Vectorization

Alves

Pestana

Silva

et al. 2016

Concurrency and Computation

View full text Add to dashboard Cite

Summary The Pre‐stack Kirchhoff Time Migration (PKTM) is a central process in petroleum exploration. As PKTM is computationally intensive, many works have proposed the use of accelerators like GPU and FPGA to improve its execution time. On the other hand, although many off‐the‐shelf processors are endowed with a set of SIMD vector instructions, few papers tackle the problem considering vectorization and all of them consider that compilers can successfully vectorize the code. In this paper, we show that programming PKTM by using SIMD vector instructions manually is more efficient than the automatically and semi‐automatically vectorized codes, provided by a hardware specific compiler and library. Experiments considering both real and synthetic datasets showed that our solution is more than four times faster than the traditional code. It also outperformed automatically vectorized codes in all tests. We believe that the proposed strategy can be used together with the other ones to accelerate seismic migration methods in general without new investments in hardware. Copyright © 2016 John Wiley & Sons, Ltd.

show abstract

“…Devidoàs inerentes características que permitem o paralelismo de dados do PKTM [Rizvandi et al 2011], muitos trabalhos como [Xu et al 2014], [Shi et al 2011], [Panetta et al 2012] e [Sun and Shi 2012] propuseram soluções para reduzir o tempo de execução deste algoritmo ao adotar dispositivos aceleradores como as Graphical Processing Units (GPU) ou Field-programable Gate Array (FPGA). Outros trabalhos como [Dai 2005] propuseram a divisão e distribuição do processamento do PKTM em cluster computacional usando o padrão MPI (Message Passing Interface). Já outros trabalhos como [Yang et al 2011] propuseram uma solução híbrida que faz uso tanto do paralelismo de memória distribuída (MPI) quanto do paralelismo de memória compartilhada (OpenMP), além de utilizar também os recursos de aceleração da GPU.…”

Section: Introductionunclassified

Vetorização e Análise de Algoritmos Paralelos para a Migração Kirchhoff Pré-empilhamento em Tempo

Silva¹,

Alves²,

Bentes³

et al. 2017

Anais Do XVIII Simpósio Em Sistemas Computacionais De Alto Desempenho (SSCAD 2017)

View full text Add to dashboard Cite

A Migração Kirchhoff Pré-empilhamento em Tempo (ou PKTM, do inglês Pre-stack Kirchhoff Time Migration) é parte central do processo de exploração de petróleo. Como o PKTM é computacionalmente intensivo, muitos trabalhos propuseram o uso de aceleradores como GPU (Graphical Processing Units) para reduzir o seu tempo de execução. Embora os processadores modernos possuam um recurso de aceleração vetorial, apenas um trabalho avaliou o uso deste recurso para acelerar o PKTM. Contudo, este trabalho avaliou apenas a versão sequencial desta aplicação. Nesse trabalho, propõe-se uma analise da vetorização de duas versões paralelas do PKTM. Para a primeira versão paralela, foi utilizado OpenMP e para a segunda, foi utilizado MPI. Em relação à vetorização, foram consideradas a vetorização automática, executada pelo compilador, e a vetorização manual, implementada pelo programador. Uma análise experimental mostrou que a vetorização automática no código com o OpenMP produziu melhores resultados do que os obtidos no código sequencial e no código com MPI. Assim, foram propostas algumas otimizações que permitiram que as versões sequenciais e com MPI obtivessem um desempenho similar ao alcançado no código com OpenMP.

show abstract

Parallel processing of Prestack Kirchhoff Time Migration on a PC Cluster

Cited by 12 publications

References 21 publications

Practical Implementation of Prestack Kirchhoff Time Migration on a General Purpose Graphics Processing Unit

Practical Implementation of Prestack Kirchhoff Time Migration on a General Purpose Graphics Processing Unit

Accelerating Pre‐stack Kirchhoff Time Migration by Manual Vectorization

Vetorização e Análise de Algoritmos Paralelos para a Migração Kirchhoff Pré-empilhamento em Tempo

Contact Info

Product

Resources

About