Using Blue Gene/P and GPUs to Accelerate Computations in the EULAG Model

详细信息    查看全文

• 作者：Roman Wyrzykowski (1) roman@icis.pcz.pl
  Krzysztof Rojek (1) krojek@icis.pcz.pl
  ?ukasz Szustak (1) lszustak@icis.pcz.pl
• 刊名：Lecture Notes in Computer Science
• 出版年：2012
• 出版时间：2012
• 年：2012
• 卷：7116
• 期：1
• 页码：670-677
• 全文大小：303.0 KB
• 参考文献：1. AMD Corporation: ATI Radeon HD 5870 Feature Summary, http://www.amd.com/
  2. Dokken, T., Hagen, T.R., Hjelmervik, J.M.: An Introduction to General-Purpose Computing on Programmable Graphics Hardware. In: Geometric Modelling, Numerical Simulation, and Optimization: Applied Mathematics at SINTEF, pp. 123–161. Springer, Heidelberg (2007)
  3. Eulag Research Model for Geophysical Flows, http://www.eulag.com/
  4. IBM Blue Gene Team: Overview of the IBM Blue Gene/P project. IBM Journal of Research and Development 52, 199–220 (2008)
  5. Khronos OpenCL Working Group: The OpenCL C++ Wrapper API, http://www.khronos.org
  6. Khronos OpenCL Working Group: The OpenCL Specification, http://www.khronos.org
  7. Lindholm, E., Nickolls, J., Oberman, S., Montrym, J.: NVIDIA Tesla: A Unified Graphics and Computing Architecture. IEEE Micro 28, 39–55 (2008)
  8. Smolarkiewicz, P., Szmelter, J.: MPDATA: An edge-based unstructured-grid formulation. Elsevier Journal of Computational Physics 206, 624–649 (2005)
  9. Sviercoski, R., Winter, C., Warrick, A.: Analytical approximation for the generalized Laplace equation with step function coefficient. J. Appl. Math. 68, 1268–1281 (2008)
  10. Tsuchiyama, R., Nakamura, N., Iizuka, T., Asahara, A., Miki, S.: The OpenCL Programming Book. Fixstars Corporation (2010)
• 作者单位：1. Czestochowa University of Technology, Dabrowskiego 73, 42-201 Czestochowa, Poland
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349

文摘

EULAG (Eulerian/semi-Lagrangian fluid solver) is an established computational model developed by the group headed by Piotr K. Smolarkiewicz for simulating thermo-fluid flows across a wide range of scales and physical scenarios. This paper presents perspectives of the EULAG parallelization based on the MPI, OpenMP, and OpenCL standards. We focus on development of computational kernels of the EULAG model. They consist of the most time-consuming calculations of the model, which are: laplacian algorithm (laplc) and multidimensional positive definite advection transport algorithm (MPDATA).