skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: One-Way Wave Equation Migration at Scale on GPUs Using Directive Based Programming

Authors:
 [1];  [1]; ORCiD logo [2];  [1]; ORCiD logo [2]
  1. Total E&P Research and Technology USA
  2. ORNL
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1399455
DOE Contract Number:
AC05-00OR22725
Resource Type:
Conference
Resource Relation:
Conference: 2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS) - Orlando, Florida, United States of America - 5/29/2017 4:00:00 AM-6/2/2017 4:00:00 AM
Country of Publication:
United States
Language:
English

Citation Formats

Mehta, Kshitij, Hugues, Maxime, Hernandez, Oscar R., Calandra, Henri, and Bernholdt, David E. One-Way Wave Equation Migration at Scale on GPUs Using Directive Based Programming. United States: N. p., 2017. Web. doi:10.1109/IPDPS.2017.82.
Mehta, Kshitij, Hugues, Maxime, Hernandez, Oscar R., Calandra, Henri, & Bernholdt, David E. One-Way Wave Equation Migration at Scale on GPUs Using Directive Based Programming. United States. doi:10.1109/IPDPS.2017.82.
Mehta, Kshitij, Hugues, Maxime, Hernandez, Oscar R., Calandra, Henri, and Bernholdt, David E. Sat . "One-Way Wave Equation Migration at Scale on GPUs Using Directive Based Programming". United States. doi:10.1109/IPDPS.2017.82. https://www.osti.gov/servlets/purl/1399455.
@article{osti_1399455,
title = {One-Way Wave Equation Migration at Scale on GPUs Using Directive Based Programming},
author = {Mehta, Kshitij and Hugues, Maxime and Hernandez, Oscar R. and Calandra, Henri and Bernholdt, David E.},
abstractNote = {},
doi = {10.1109/IPDPS.2017.82},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Sat Jul 01 00:00:00 EDT 2017},
month = {Sat Jul 01 00:00:00 EDT 2017}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Graphics Processing Unit (GPU)-based parallel computer architectures have shown increased popularity as a building block for high performance computing, and possibly for future Exascale computing. However, their programming complexity remains as a major hurdle for their widespread adoption. To provide better abstractions for programming GPU architectures, researchers and vendors have proposed several directive-based GPU programming models. These directive-based models provide different levels of abstraction, and required different levels of programming effort to port and optimize applications. Understanding these differences among these new models provides valuable insights on their applicability and performance potential. In this paper, we evaluate existing directive-based modelsmore » by porting thirteen application kernels from various scientific domains to use CUDA GPUs, which, in turn, allows us to identify important issues in the functionality, scalability, tunability, and debuggability of the existing models. Our evaluation shows that directive-based models can achieve reasonable performance, compared to hand-written GPU codes.« less
  • Directive-based, accelerator programming models such as OpenACC have arisen as an alternative solution to program emerging Scalable Heterogeneous Computing (SHC) platforms. However, the increased complexity in the SHC systems incurs several challenges in terms of portability and productivity. This paper presents an open-sourced OpenACC compiler, called OpenARC, which serves as an extensible research framework to address those issues in the directive-based accelerator programming. This paper explains important design strategies and key compiler transformation techniques needed to implement the reference OpenACC compiler. Moreover, this paper demonstrates the efficacy of OpenARC as a research framework for directive-based programming study, by proposing andmore » implementing OpenACC extensions in the OpenARC framework to 1) support hybrid programming of the unified memory and separate memory and 2) exploit architecture-specific features in an abstract manner. Porting thirteen standard OpenACC programs and three extended OpenACC programs to CUDA GPUs shows that OpenARC performs similarly to a commercial OpenACC compiler, while it serves as a high-level research framework.« less
  • A first-order one-way wave system has been created based on characteristic analysis of the acoustic wave system and optimization of the dispersion relation. The authors demonstrate that this system is equivalent to a third-order scalar partial-differential equation which, for a homogeneous medium, reduces to a form similar to the 45{degree} paraxial wave equation. This system describes accurately waves propagating in a 2D heterogeneous medium at angles up to 75{degree}. The one-way wave system representing downgoing waves is used for a modified reverse time migration method. As a wavefield extrapolator in migration, the downgoing wave system propagates the reflection events backwardsmore » to their reflectors without scattering at the discontinuities in the velocity model. Hence, images with amplitudes proportional to reflectivity can be obtained from this migration technique. They present examples of the application of the new migration method to synthetic seismic data where P-P reflections P-SV converted waves are present. Absorbing boundaries, useful in the generation of synthetic seismograms, have been constructed by using the one-way wave system. These boundaries absorb effectively waves impinging over a wide range of angles of incidence.« less
  • In this paper the authors present a method for constructing one-way wave equation with the use of a parameter that is related to the angle of wave incidence. The constructed wave equation can be satisfied for plane incident waves at two directions, and split into two low order equations similar to 15{degrees} or 45{degrees} equation respectively. By identifying the near normal incident and near glancing waves, and giving the three values {theta} = 0, {pi}/4, and {pi}/2, we can effectively resolve three problems: the incorrectness in instantaneous determination of incident angle, poor posed boundary and boundary reflection. The boundary treatmentsmore » of both stack section made from plane wave sources and point-shot records prove this method excellent. The application of such one-way wave equation to exploiting WIMIG module for migration is described fully. This application would be useful for the petroleum industry.« less