Abstract
High performance computing systems are increasingly incorporating the computational power provided by accelerators, especially GPUs. With the programmability of GPUs greatly facilitated by OpenCL or NVIDIA’s CUDA, with support for full double precision on GPUs, many challenging problems are benefiting from these processing units. It is well-known that memory latency is the speed limiting factor on GPUs. To hide memory latency, kernel instances must be executed in parallel on the same core, making sparse data more difficult to deal with than dense data. In this work we examine the numerical solution of a radiative transfer problem. We show that integral problem formulations relying on sparse linear algebra computations can benefit from the computing power of such devices, achieving an average speedup of 50× when compared to a representative CPU implementation.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Abramowitz, M., Stegun, I.A.: Handbook of Mathematical Functions. Dover, New York (1960)
Ahues, M., d’Almeida, F.D., Largillier, A., Titaud, O., Vasconcelos, P.: An L 1 refined projection approximate solution of the radiation transfer equation in stellar atmospheres. Journal of Computational and Applied Mathematics 140(1-2), 13–26 (2002)
Bell, N., Dalton, S., Olson, L.N.: Exposing fine-grained parallelism in algebraic multigrid methods. SIAM Journal on Scientific Computing 34(4), C123 – C152 (2012)
Bell, N., Garland, M.: Cusp: Generic parallel algorithms for sparse matrix and graph computations (2012), http://cusp-library.googlecode.com , version 0.3.0
Bell, N., Hoberock, J.: Thrust: A productivity-oriented library for CUDA. GPU Computing Gems 7 (2011)
Benzi, M.: Preconditioning techniques for large linear systems: a survey. Journal of Computational Physics 182(2), 418–477 (2002)
Busbridge, I.: The mathematics of radiative transfer, vol. 369. Cambridge University Press (1960)
Dongarra, J., Beckman, P., Moore, T., Aerts, P., Aloisio, G., Andre, J., Barkai, D., Berthou, J., Boku, T., Braunschweig, B., et al.: The international exascale software project roadmap. International Journal of High Performance Computing Applications 25(1), 3–60 (2011)
NVIDIA Corporation: CULA Sparse Reference Manual (2013), http://www.culatools.com/cula_sparse_programmers_guide/
NVIDIA Corporation: NVIDIA CUDA C Programming Guide (February 2014), http://docs.nvidia.com/cuda/
Roman, J.E., Vasconcelos, P.B.: Harnessing GPU power from high-level libraries: Eigenvalues of integral operators with SLEPc. Procedia Computer Science 18, 2591–2594 (2013)
Rupp, K., Rudolf, F., Weinbub, J.: ViennaCL - A High Level Linear Algebra Library for GPUs and Multi-Core CPUs. In: Intl. Workshop on GPUs and Scientific Applications, pp. 51–56 (2010)
Rutily, B.: Multiple scattering theory and integral equations. In: Constanda, C., Ahues, M., Largillier, A. (eds.) Integral Methods in Science and Engineering, pp. 211–232. Birkhäuser (2004)
Rutily, B., Chevallier, L.: The finite Laplace transform for solving a weakly singular integral equation occurring in transfer theory. Journal of Integral Equations and Applications 16(4), 389–409 (2004)
Saad, Y., van der Vorst, H.A.: Iterative solution of linear systems in the 20th century. Journal of Computational and Applied Mathematics 123, 1–33 (2000)
Stone, J.E., Gohara, D., Shi, G.: OpenCL: A parallel programming standard for heterogeneous computing systems. IEEE Design & Test 12(3), 66–73 (2010), http://dx.doi.org/10.1109/MCSE.2010.69
Tuminaro, R.S.: Parallel smoothed aggregation multigrid: aggregation strategies on massively parallel machines. In: Proceedings of the 2000 ACM/IEEE conference on Supercomputing (CDROM). p. 5. IEEE Computer Society (2000)
Vasconcelos, P.B., d’Almeida, F.: Performance evaluation of a parallel algorithm for a radiative transfer problem. In: Dongarra, J., Madsen, K., Waśniewski, J. (eds.) PARA 2004. LNCS, vol. 3732, pp. 864–871. Springer, Heidelberg (2006)
Vasconcelos, P.B., Marques, O., Roman, J.E.: Parallel eigensolvers for a discretized radiative transfer problem. In: Palma, J.M.L.M., Amestoy, P.R., Daydé, M., Mattoso, M., Lopes, J.C. (eds.) VECPAR 2008. LNCS, vol. 5336, pp. 336–348. Springer, Heidelberg (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Vasconcelos, P.B., Marques, O. (2014). Experiments with GPU-Acceleration for Solving a Radiative Transfer Problem. In: Murgante, B., et al. Computational Science and Its Applications – ICCSA 2014. ICCSA 2014. Lecture Notes in Computer Science, vol 8582. Springer, Cham. https://doi.org/10.1007/978-3-319-09147-1_40
Download citation
DOI: https://doi.org/10.1007/978-3-319-09147-1_40
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-09146-4
Online ISBN: 978-3-319-09147-1
eBook Packages: Computer ScienceComputer Science (R0)