Accelerating the Sweep3D for a Graphic Processor Unit
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
As a powerful and flexible processor, the Graphic Processing Unit (GPU) can offer a great faculty in solving many high-performance computing applications. Sweep3D, which simulates a single group time-independent discrete ordinates (Sn) neutron transport deterministically on 3D Cartesian geometry space, represents the key part of a real ASCI application. The wavefront process for parallel computation in Sweep3D limits the concurrent threads on the GPU. In this paper, we present multi-dimensional optimization methods for Sweep3D, which can be efficiently implemented on the fine-grained parallel architecture of the GPU. Our results show that the overall performance of Sweep3D on the CPU-GPU hybrid platform can be improved up to 4.38 times as compared to the CPU-based implementation.
Abstract
As a powerful and flexible processor, the Graphic Processing Unit (GPU) can offer a great faculty in solving many high-performance computing applications. Sweep3D, which simulates a single group time-independent discrete ordinates (Sn) neutron transport deterministically on 3D Cartesian geometry space, represents the key part of a real ASCI application. The wavefront process for parallel computation in Sweep3D limits the concurrent threads on the GPU. In this paper, we present multi-dimensional optimization methods for Sweep3D, which can be efficiently implemented on the fine-grained parallel architecture of the GPU. Our results show that the overall performance of Sweep3D on the CPU-GPU hybrid platform can be improved up to 4.38 times as compared to the CPU-based implementation.
참고문헌 (22)
- H. Nguyen, "GPU Gems 3" (2007), Addison Wesley
- D. Kirk, "Innovation in graphics technology", Canadian Undergraduate Technology Conference (2004)
- "ATI Radeon HD 5870 Feature Summary" (2010)[URL]
- NVIDIA Corporation, "CUDA Programming Guide Version 3.1" (2010)
- AMD Corporation, "ATI Stream Computing User Guide Version 2.0" (2010)
- A. Munshi, "The OpenCL Specification Version: 1.1" (2010), Khronos OpenCL Working Group
- "Vertical solutions on CUDA" (2010)[URL]
- M.M. Mathis, "A General Performance Model for Parallel Sweeps on Orthogonal Grids for Particle Transport Calculations", Proc. ACM Int. Conf. Supercomputing, pp.255~263 (2000)
- A. Hoisie, "Scalability analysis of multidimensional wavefront algorithms on large-scale SMP clusters", The 7th Symposium on the Frontiers of Massively Parallel Computation, pp.4~15 (1999)
- A. Hoisie, "Performance and scalability analysis of teraflop- scale parallel architectures using multidimensional wavefront applications", International Journal of High Performance Computing Applications, 14(4), pp.330~346 (2000)
- "Sweep3D" (2010)[URL]
- K. Davis, "A Performance and Scalability Analysis of the BlueGene", L Architecture, IEEE conference on Supercomputing, pp.2004[DOI]
- K.J. Barker, "Entering the petaflop era: the architecture and performance of Roadrunner", Proceedings of the 2008 ACM, 2008, 1~11
- E.E. Lewis, "Computational Methods of Neutron Transport" (1993), American Nuclear Society
- K. Koch, "Solution of the First-Order Form of Three-Dimensional Discrete Ordinates Equations on a Massively Parallel Machine", Transactions of American Nuclear Society, 65, pp.198~199 (1992)
- M.M Mathis, "A General Performance Model of structured and Unstructured Mesh Particle Transport Computations", Journal of Supercomputing, 34, pp.181~199 (2005)
- D.J. Kerbyson, "Analysis of Wavefront Algorithms on Large-scale Two level Heterogene ous Processing Systems", Workshop on Unique Chips and Systems, pp.259~279 (2006)
- F. Petrini, "Multicore Surprises: Lessons Learned from Optimizing Sweep3D on the Cell Broadband Engine", The 21st International Parallel and Distributed Processing Symposium (2007)
- "NVIDIA Tesla S1070 1U Computing System" (2010)[URL]
- V.Volkov, "Benchmarking GPUs to tune dense linear algebra", Proceedings of the 2008 ACM, 2008, ~
- O. Lubeck, "Implementation and performance modeling of deterministic particle transport (Sweep3D) on the IBM Cell", BE, 2009, pp.17(1), ~
- C. Gong, "Optimizing Sweep3D for Graphic Processor Unit, In Vol.6081 of Lecture Notes in Computer Science", pp.416~426 (2010), Springer