Abstract
A parallel algorithm is derived for solving the discrete-ordinates approximation of the neutron transport equation, based on the naturally occurring decoupling between the mesh sweeps in the various discrete directions during each iteration. In addition, the parallel Source Iteration (SI) algorithm, characterized by its coarse granularity and static scheduling, is implemented for the Nodal Integral Method (NIM) into the Parallel Nodal Transport (P-NT) code on Intel's iPSC/2 hypercube. Measured parallel performance for solutions of two test problems is used as evidence of the parallel algorithm's potential for high speedup and efficiency. The measured performance data are also used to develop and validate a parallel performance model for the total, serial, parallel, and global-summation time components per iteration as a function of the spatial mesh size, the problem size (number of mesh cells and angular quadrature order), and the number of utilized processors. The potential for high performance (large speedup at high efficiency) for large problems is explored using the performance model, and it is concluded that present applications in three-dimensional Cartesian geometry will benefit by concurrent execution on parallel computers with up to a few hundred processors.
Similar content being viewed by others
References
Azmy, Y.Y. 1988a. Comparison of three approximations to the linear-linear nodal transport method in weighted diamond-difference form. Nuclear Science & Engineering, 100: 190.
Azmy, Y.Y. 1988b. Multidimensional nodal transport methods for multiple-instruction multiple-data, distributed memory machines. Trans., Am. Nuclear Soc., 56: 292.
Azmy, Y.Y. 1988c. The weighted diamond-difference form of nodal transport methods. Nuclear Science & Engineering, 98: 29.
Azmy, Y.Y. 1990a. On the adequacy of message-passing parallel supercomputers for solving neutron transport problems. In Proc., Supercomputing '90, IEEE Comp. Soc. Press, p. 693.
Azmy, Y.Y. 1990b. Parallel performance models for the nodal transport method on the hypercube. Trans., Am. Nuclear Soc., 62: 300.
Azmy, Y.Y. 1992. General order nodal transport methods and application to parallel computing. Transport Theory and Statistical Physics (to appear).
Bell, G.I., and Glasstone, S. 1970. Nuclear Reactor Theory. Van Nostrand Reinhold, New York.
Carlson, B.G., and Lathrop, K.D. 1968. Transport-theory—The method of discrete ordinates. In Computing Methods (H. Greenspan, C.N. Kelber, and D. Okrent, eds.), Gordon and Breach, New York.
Dorning, J.J. 1985. Nodal transport methods after five years. In Proc., Topical Meeting on Advances in Nuclear Engineering Computational Methods, vol. 2 (Knoxville, Tenn., Apr. 9–11), p. 412.
Geist, G.A., Heath, M.T., Peyton, B.W., and Worley, P.H. 1990. A user's guide to PICL: A Portable Instrumented Communication Library. ORNL/TM-11616, Oak Ridge Nat. Laboratory, Oak Ridge, Tenn.
Haghighat, A., and Azmy, Y.Y. 1991. Parallelization of a spherical S N algorithm based on the spatial domain decomposition. In Proc., ANS Topical Meeting on Advances in Mathematics, Computations, and Reactor Physics, 1.1 3 (Pittsburgh, Apr. 28–May 2), Am. Nuclear Soc.
Haghighat, A., and Mattis, R. 1990. Parallel/vector algorithms for the spherical Sn transport theory method. In Proc., Internat. Conf. on Supercomputing in Nuclear Applications (Mito City, Ibaraki, Japan, Mar. 12–19), p. 295.
Hanebutte, U.R., and Lewis, E.E. 1991. A discrete ordinate response matrix method for massively parallel computers. In Proc., ANS Topical Meeting on Advances in Mathematics, Computations, and Reactor Physics, 1.1 1 (Pittsburgh, Apr. 28–May 2), Am. Nuclear Soc.
Heath, M.T. 1990. Visual animation of parallel algorithms for matrix computations. In Proc., Fifth Distributed Memory Computing Conf. (D. Walker, ed.).
Lawrence, R.D. 1986. Progress in nodal methods for the solution of the neutron diffusion and transport equations. Prog. Nuclear Energy, 17: 271.
Lewis, E.E., and Miller, W.F. 1984. Computational Methods of Neutron Transport. John Wiley and Sons, New York.
Mattis, R.E., and Haghighat, A. 1991. New vector/parallel algorithms for two-dimensional curvilinear SN method. In Proc., ANS Topical Meeting on Advances in Mathematics, Computations, and Reactor Physics, 1.1 4 (Pittsburgh, Apr. 28–May 2), Am. Nuclear Soc.
Morel, J.E., McGhee, J.M., Olvey, L.A., and Claborn, G.W. 1991. Solution of the 3-D even-parity S N equations on the Connection Machine. In Proc., ANS Topical Meeting on Advances in Mathematics, Computations, and Reactor Physics, 1.1 2 (Pittsburgh, Apr. 28–May 2), Am. Nuclear Soc.
Rhoades, W.A., and Childs, R.L. 1987. Sn transport calculations on vector and parallel processors. Trans., Am. Nuclear Soc., 55: 320.
Rhoades, W.A., and Flanery, R.E. 1989. 3-D discrete ordinates calculations with parallel-vector processors. In Proc., ANS Topical Meeting on Advances in Nuclear Engineering Computation and Radiation Shielding, 69 (Santa Fe, N. Mex., Apr. 9–13), Am. Nuclear Soc.
Wienke, B.R., and Hiromoto, R.E. 1985. Parallel Sn iteration schemes. Nuclear Science & Engineering, 90: 116.
Yavuz, M., and Larsen, E.W. 1989. Spatial domain decomposition methods for discrete-ordinates problems. In Proc., ANS Topical Meeting on Advances in Nuclear Engineering Computation and Radiation Shielding, 12 (Santa Fe, N. Mex., Apr. 9–13), Am. Nuclear Soc.
Author information
Authors and Affiliations
Additional information
Research sponsored by the U.S. Department of Energy, managed by Martin Marietta Energy Systems, Inc., under contract No. DE-AC05-84OR21400.
Rights and permissions
About this article
Cite this article
Azmy, Y.Y. Performance and performance modeling of a parallel algorithm for solving the neutron transport equation. J Supercomput 6, 211–235 (1992). https://doi.org/10.1007/BF00155800
Issue Date:
DOI: https://doi.org/10.1007/BF00155800