The Community for Technology Leaders
RSS Icon
Issue No.01 - Jan.-Feb. (2013 vol.15)
pp: 46-55
To reduce the design complexity of OpenCL programming, the approach proposed here generates application code automatically, based on model-driven engineering (MDE) and modeling and analysis of real-time and embedded (MARTE) systems. The aim is to provide application-development resources for nonspecialists in parallel programming, exploiting concepts such as reuse and platform independence.
Unified modeling language, Computational modeling, Computer architecture, Software engineering, Resource management, Parallel programming, Scientific computing, scientific computing, model-driven engineering, MARTE, OpenCL, GPU, high-performance computing
A. Wendell O. Rodrigues, Frederic Guyomarc'h, Jean-Luc Dekeyser, "An MDE Approach for Automatic Code Generation from UML/MARTE to OpenCL", Computing in Science & Engineering, vol.15, no. 1, pp. 46-55, Jan.-Feb. 2013, doi:10.1109/MCSE.2012.35
1. Khronos OpenCL Working Group, The OpenCL Specification, version 1.1, revision 44, 2011.
2. D. Lugato, J.-M. Bruel, and I. Ober, “Model-Driven Engineering for High Performance Computing Applications,” Proc. Modeling Simulation and Optimization Focus on Applications, Acta Press, 2010, pp. 303–308.
3. A. Gamatié et al., “A Model Driven Design Framework for Massively Parallel Embedded Systems,” ACM Trans. Embedded Computing Systems, vol. 10, no. 4, 2011, article no. 39.
4. C. Glitia et al., Repetitive Model Refactoring for Design Space Exploration of Intensive Signal Processing Applications, tech. report, INRIA, 2009.
5. P. Boulet, Array-OL Revisited, Multidimensional Intensive Signal Processing Specification, tech. report, INRIA, 2007.
6. W. Rodrigues, “A Methodology to Develop High Performance Applications on GPGPU Architectures: Application to Simulation of Electrical Machines,” doctoral thesis, Computer Science Dept., Univ. des Sciences et Technologie de Lille, 2012.
7. G.H. Golub and C.F. Van Loan, Matrix Computations, 3rd ed., The Johns Hopkins Univ. Press, 1996.
8. W. Rodrigues et al., “Automatic MultiGPU Code Generation Applied to Simulation of Electrical Machines,” IEEE Trans. Magnetics, vol. 48, no. 2, 2012, pp. 831–834.
9. A. Cevahir, A. Nukada, and S. Matsuoka, “High Performance Conjugate Gradient Solver on Multi-GPU Clusters Using Hypergraph Partitioning,” Computer Science Research and Development, vol. 25, nos. 1–2, 2010, pp. 83–91.
24 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool