Abstract
Monte Carlo simulation is an important method of uncertainty quantification in a variety of scientific and engineering disciplines. Computational scalability is a critical problem in Monte Carlo implementation for many dynamical systems. However, new high-throughput computing architectures such as GPU’s are enabling realizations of Monte Carlo that exhibit 1–2 orders-of-magnitude runtime reductions over comparable serial methods. This chapter outlines the basic process of Monte Carlo implementation on GPU’s including optimization techniques and cost-benefit considerations. Example source code listings are provided for Monte Carlo simulation of a point-mass in atmospheric flight. Runtime results are provided demonstrating significant runtime reductions compared to serial CPU-based implementations.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Dunn, W.L.: Exploring Monte Carlo Methods. Elsevier, Amsterdam (2012)
Moler, C.: Matrix computation on distributed memory multiprocessors. In: Hypercube Multiprocessors. Society for Industrial and Applied Mathematics, Philadelphia (1986). ISBN 0898712092
Liu, T., Ding, A., Xu, X.: GPU-based Monte Carlo methods for accelerating radiographic and CT imaging dose calculations: feasibility and scalability. Med. Phys. 39(6), 3876 (2012)
Humphrey, A., Meng, Q., Berzins, M., Harman, T.: Radiation modeling using the Uintah heterogeneous CPU/GPU runtime system. In: Proceedings of the First Conference of the Extreme Science and Engineering Discovery Environment (XSEDE’12), vol. 4, pp. 4.1–4.8 (2012)
Rogers, J., Slegers, N.: Robust parafoil terminal guidance using massively parallel processing. J. Guid. Control. Dyn. 36(5), 1336–1345 (2013)
Rogers, J.: GPU-Enabled Projectile Guidance for Impact Area Constraints. In: SPIE Defense, Security, and Sensing Symposium, Baltimore (2013)
Kolb, C., Pharr. M.: Option pricing on the GPU. GPU Gems 2, NVIDIA Corporation, Chapter 45 (2005)
CUDA Toolkit 4.2: CURAND Guide. NVIDIA Corporation, Santa Clara (2012)
CUDA C Programming Guide.: NVIDIA Corporation, Santa Clara (2012)
Ilg, M., Rogers, J., Costello, M.: Projectile Monte Carlo analysis using a graphics processing unit. In: AIAA Atmospheric Flight Mechanics Conference, Portland (2011)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Rogers, J. (2014). Monte Carlo Simulation of Dynamic Systems on GPU’s. In: Kindratenko, V. (eds) Numerical Computations with GPUs. Springer, Cham. https://doi.org/10.1007/978-3-319-06548-9_15
Download citation
DOI: https://doi.org/10.1007/978-3-319-06548-9_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-06547-2
Online ISBN: 978-3-319-06548-9
eBook Packages: Computer ScienceComputer Science (R0)