Full Article - Open Access.

Idioma principal


Taniguchi, D.; Sato, L. M.; Cheng, L. Y.;

Full Article:

Moving Particle Semi-implicit (MPS) is a Lagrangian, meshfree, computational method for fluid simulation. This work focus on using GPU clusters for MPS simulations. To accomplish this, we have to deal with two different levels of parallelism: one responsible for making different cluster nodes work together in a distributed memory system, and the other using the parallelism of GPU devices available on each node. First we present a performance comparison between single-node GPU and single-node multithreaded CPU implementations to investigate GPU speedups. Further, we analyze the performance in a multi-node GPU cluster environment.

Full Article:

Palavras-chave: Particle Method, Computational Fluid Dynamics, High Performance Computing, GPU, CUDA.,


DOI: 10.5151/meceng-wccm2012-18291

Referências bibliográficas
  • [1] Aji A. M., Zhang L., Feng W. C., “GPU-RMAP: accelerating short-read mapping on graphics processors”. Proc. IEEE Conf. Computational Science and Engineering, 168-175, 201
  • [2] Barnes J., Hut P., “A hierarchical O(nlogn) force-calculation algorithm”. Nature, 324, 446-449, 1986.
  • [3] Barnes J. E., “A modified tree code: don’t laugh; it runs”. J. of Comput. Phys., 87, 161-170, 1990.
  • [4] Cole R., “Underwater explosions”. Princeton University Press, 1948.
  • [5] Friedrichs M. S., et al., “Accelerating molecular dynamic simulation on graphics processing units”. J. Comput. Chem., 30, 864-872, 2009.
  • [6] Garland M., et al., “Parallel Computing Experiences with CUDA”. IEEE Micro, 28, 13-27, 2008.
  • [7] Green S., “Particle simulation using CUDA”. NVIDIA Whitepaper, 2010.
  • [8] Hamada T., Nitadori K., “190 TFlops astrophysical n-body simulation on a cluster of GPUs”. Proc. IEEE Conf. for High Performance Computing, 1-9, 2010.
  • [9] Isshiki H., “Discrete differential operators on irregular nodes (DDIN)”. Int. J. Num. Meth. Eng., 88, 1323-1343, 2011.
  • [10] Koshizuka S., Oka Y., “Moving particle semi-implicit method for fragmentation of in-compressible fluid”. Nuclear Science and Engineering, 123, 421-434, 1996.
  • [11] Lee M., Jeon J. H., Kim J., Song J., “Scalable and parallel implementation of a financial application on a GPU: with focus on out-of-core case.” Proc. IEEE Conf. Computer and Information Technology, 1323-1327, 2010.
  • [12] Le Grand S., “Broad-phase collision detection with CUDA”. GPU Gems 3, 697-721, 2008.
  • [13] Liu Y., Zhang E. Z., Shen X., “A cross-input adaptive framework for GPU program optimizations”. Proc. IEEE Symp. Parallel Andamp; Distributed Processing, 1-10, 2009.
  • [14] Monaghan J. J., “Simulating free surface flows with SPH”. J. Comput. Phys., 110, 399-406, 1994.
  • [15] NVIDIA CUDA, NVIDIA CUDA C Programming Guide, 4.1, 2011.
  • [16] Oochi M., Yamada Y., Koshizuka S., Sakai M., “Explicit MPS algorithm for free sur-face flow analysis”. Trans. of the Japan Society for Computational Engineering and Science, 2011.
  • [17] Shakibaeinia A., Jin Y. C., “A weakly compressible MPS merhod for modeling of open-boundary free-surface flow”. Int. J. Numer. Meth. Fluids, 63, 1208-1232, 2010.
Como citar:

Taniguchi, D.; Sato, L. M.; Cheng, L. Y.; "EXPLICIT MOVING PARTICLE SIMULATION METHOD ON GPU CLUSTERS", p. 1155-1168 . In: In Proceedings of the 10th World Congress on Computational Mechanics [= Blucher Mechanical Engineering Proceedings, v. 1, n. 1]. São Paulo: Blucher, 2014.
ISSN 2358-0828, DOI 10.5151/meceng-wccm2012-18291

últimos 30 dias | último ano | desde a publicação