FPGA与GPU并行计算分析——以Kirchhoff叠前时间偏移为例
|
摘要
高性能计算(high performance computing)是地震数据处理领域一个重要研究课题.目前的高性能计算大多是利用图形处理器(GPU)强大并行处理能力,以GPU/CPU协同并行计算构架为基础实现有关算法.本文讨论了利用CUDA和Impulse C-to-FPGA工具分别在GPU(图形处理器)和FPGA(可编程逻辑门阵列)平台上并行实现Kirchhoff叠前时间偏移算法.处理结果表明,利用GPU加速Kirchhoff叠前时间偏移处理得到了14倍的单核加速比,利用FPGA加速Kirchhoff叠前时间偏移处理得到了9.5倍的单核加速比.
High performance computing is an important research subject in the area of seismic data processing.At present,most of the high performance computing is to use the powerful parallel processing ability of GPU and implement the algorithms on the basis of GPU/CPU co-processing parallel system.The paper discusses parallel implementation of prestack Kirchhoff time migration algorithm on GPU(Graphic Processing Unit) using CUDA,as well as on FPGA(Field-Programmable Gate Array) with the Impulse C-to-FPGA tool.Results show that GPU implementation achieves approximately 14 speedup,and FPGA implementation achieves about 9.5 speedup.
High performance computing is an important research subject in the area of seismic data processing.At present,most of the high performance computing is to use the powerful parallel processing ability of GPU and implement the algorithms on the basis of GPU/CPU co-processing parallel system.The paper discusses parallel implementation of prestack Kirchhoff time migration algorithm on GPU(Graphic Processing Unit) using CUDA,as well as on FPGA(Field-Programmable Gate Array) with the Impulse C-to-FPGA tool.Results show that GPU implementation achieves approximately 14 speedup,and FPGA implementation achieves about 9.5 speedup.
引文
[1]赵改善.可重构计算技术及其在地球物理中的应用前景[J].勘探地球物理进展,2007,30(4):309-316.Zhao G S.Reconfigurable computing and its applicationprospects in geophysical exploration[J].Progress inExploration Geophysics(in Chinese),2007,30(4):309-316.
[2]黄易,师学明,范建柯,等.并行计算技术及其在勘探地球物理学中的现状与展望[J].地球物理学进展,2010,25(2):642-649.Huang Y,Shi X M,Fan J K,et al.Review on parallelcomputing and its application in exploration geophysics[J].Progress in Geophys.(in Chinese),2010,25(2):642-649.
[3]Panetta J,Teixeira T,de Souza Filho P R P,et al.AcceleratingKirchhoff migration by CPU and GPU cooperation[A].//Proceedings of the 21st International Symposium on ComputerArchitecture and High Performance Computing[C].SaoPaulo:IEEE,2009:26-32.
[4]Shi X H,Li C,Wang X,et al.A practical approach of curvedray prestack Kirchhoff time migration on GPGPU[C].Advanced Parallel Processing Technologies Lecture Notes inComputer Science,2009,165-176.
[5]Micikevicius P.3Dfinite difference computation on GPUsusing CUDA[C].Proceedings of 2nd Workshop on GeneralPurpose Processing on Graphics Processing Units,2009,79-84.
[6]William B.A novel GPGPU approach to kirchhoff time migration[C].SEG Annual Meeting,2011.
[7]He C,Lu M,Sun C W.Accelerating seismic migration usingFPGA-based coprocessor platform[C].Proc.FCCM,2004,207-216.
[8]张栋,杜清怀,许孝坤.高性能FPGA处理器在叠前可希霍夫偏移中的应用[J].西北地质,2004,37(4):102-106.Zhang D,Du Q H,Xu X K.High Performance FPGA-basedcoprocessor platform for accelerating Kirchhoff migration[J].Northwestern Geology(in Chinese),2004,37(4):102-106.
[9]王棣,王华忠,马在田,等.叠前时间偏移方法综述[J].勘探地球物理进展,2004,27(5):313-320.Wang D,Wang H Z,Ma Z T,et al.Review of Prestack timemigration methods[J].Progress in Exploration Geophysics(inChinese),2004,27(5):313-320.
[10]李肯立,彭俊杰,周仕勇.基于CUDA的Kirchhoff叠前时间偏移算法设计与实现[J].计算机应用研究,2009,26(12):4474-4477.Li K L,Peng J J,Zhou S Y.Implement Kirchhoff prestacktime migration algorithm on CUDA architecture[J].Application Research of Computers(in Chinese),2009,26(12):4474-4477.
[11]Castle R J.Wave-equation migration in the presence of lateralvelocity variations[J].Geophysics,1982,47(7):1001-1011.
[12]Yilmaz O.Pre-stack partial migration[Ph.D.thesis].Stanford:Stanford University,1979.
[13]Lumley D E,Claerbout J F,Bevc D.Anti-aliased Kirchhoff3-D migration[C].64th Ann.Internat.Mtg.,Soc.Expl.Geophys.Expanded Abstracts,1994,1282-1285.
[14]Yilmaz O,Claerbout J F.Prestack partial migration[J].Geophysics,1980,45(12):1753-1779.
[15]Yilmaz O.Seismic data analysis[M].Tulsa:Society ofExploration Geophysicists,2001.
[16]Claerbou J F.Imaging the earth’s interior[M].Cambridge:Blackwell Scientific Publications,1985.
[17]李博,刘国峰,刘洪.地震叠前时间偏移的一种图形处理器提速实现方法[J].地球物理学报,2009,52(1):245-252.Li B,Liu G F,Liu H.A method of using GPU to accelerateseismic pre-stack time migration[J].Chinese J.Geophys.(in Chinese),2009,52(1):245-252.
[18]NVIDIA CUDA.programming Guide Version 3.1,http://developer.download.nvidia.com/.
[19]David P,Scott T.Practical FPGA programming in C[M].Boston:Prentice Hall PTR,2005.
[20]Impulse Acclerate Technologies.ImpulseC User’x Guide.http://www.impulseaccelerated.com/.
[21]Impulse Acclerate Technologies.Codeveloper User’x Guide.http://www.impulseaccelerated.com/.
[2]黄易,师学明,范建柯,等.并行计算技术及其在勘探地球物理学中的现状与展望[J].地球物理学进展,2010,25(2):642-649.Huang Y,Shi X M,Fan J K,et al.Review on parallelcomputing and its application in exploration geophysics[J].Progress in Geophys.(in Chinese),2010,25(2):642-649.
[3]Panetta J,Teixeira T,de Souza Filho P R P,et al.AcceleratingKirchhoff migration by CPU and GPU cooperation[A].//Proceedings of the 21st International Symposium on ComputerArchitecture and High Performance Computing[C].SaoPaulo:IEEE,2009:26-32.
[4]Shi X H,Li C,Wang X,et al.A practical approach of curvedray prestack Kirchhoff time migration on GPGPU[C].Advanced Parallel Processing Technologies Lecture Notes inComputer Science,2009,165-176.
[5]Micikevicius P.3Dfinite difference computation on GPUsusing CUDA[C].Proceedings of 2nd Workshop on GeneralPurpose Processing on Graphics Processing Units,2009,79-84.
[6]William B.A novel GPGPU approach to kirchhoff time migration[C].SEG Annual Meeting,2011.
[7]He C,Lu M,Sun C W.Accelerating seismic migration usingFPGA-based coprocessor platform[C].Proc.FCCM,2004,207-216.
[8]张栋,杜清怀,许孝坤.高性能FPGA处理器在叠前可希霍夫偏移中的应用[J].西北地质,2004,37(4):102-106.Zhang D,Du Q H,Xu X K.High Performance FPGA-basedcoprocessor platform for accelerating Kirchhoff migration[J].Northwestern Geology(in Chinese),2004,37(4):102-106.
[9]王棣,王华忠,马在田,等.叠前时间偏移方法综述[J].勘探地球物理进展,2004,27(5):313-320.Wang D,Wang H Z,Ma Z T,et al.Review of Prestack timemigration methods[J].Progress in Exploration Geophysics(inChinese),2004,27(5):313-320.
[10]李肯立,彭俊杰,周仕勇.基于CUDA的Kirchhoff叠前时间偏移算法设计与实现[J].计算机应用研究,2009,26(12):4474-4477.Li K L,Peng J J,Zhou S Y.Implement Kirchhoff prestacktime migration algorithm on CUDA architecture[J].Application Research of Computers(in Chinese),2009,26(12):4474-4477.
[11]Castle R J.Wave-equation migration in the presence of lateralvelocity variations[J].Geophysics,1982,47(7):1001-1011.
[12]Yilmaz O.Pre-stack partial migration[Ph.D.thesis].Stanford:Stanford University,1979.
[13]Lumley D E,Claerbout J F,Bevc D.Anti-aliased Kirchhoff3-D migration[C].64th Ann.Internat.Mtg.,Soc.Expl.Geophys.Expanded Abstracts,1994,1282-1285.
[14]Yilmaz O,Claerbout J F.Prestack partial migration[J].Geophysics,1980,45(12):1753-1779.
[15]Yilmaz O.Seismic data analysis[M].Tulsa:Society ofExploration Geophysicists,2001.
[16]Claerbou J F.Imaging the earth’s interior[M].Cambridge:Blackwell Scientific Publications,1985.
[17]李博,刘国峰,刘洪.地震叠前时间偏移的一种图形处理器提速实现方法[J].地球物理学报,2009,52(1):245-252.Li B,Liu G F,Liu H.A method of using GPU to accelerateseismic pre-stack time migration[J].Chinese J.Geophys.(in Chinese),2009,52(1):245-252.
[18]NVIDIA CUDA.programming Guide Version 3.1,http://developer.download.nvidia.com/.
[19]David P,Scott T.Practical FPGA programming in C[M].Boston:Prentice Hall PTR,2005.
[20]Impulse Acclerate Technologies.ImpulseC User’x Guide.http://www.impulseaccelerated.com/.
[21]Impulse Acclerate Technologies.Codeveloper User’x Guide.http://www.impulseaccelerated.com/.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心