Abstract
The GPU-native CFD framework with dynamical adaptive mesh refinement (AMR) requires periodical execution of memory compaction operations to relieve memory expenses. The present paper addresses several different parallel GPU memory defragmentation algorithms for octree-based AMR grids. These algorithms are tested on benchmark CFD problems typical for AMR transformation. The results show that the memory defragmentation algorithm based on the prefix scan procedure is not only 1–2 order faster compared to algorithm based on space filling curve (z-curve) but also surprisingly and dramatically impacts on CFD solver performance by reducing total GPU runtime for some problems up to 37%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Pavlukhin, P., Menshov, I.: On implementation high-scalable CFD solvers for hybrid clusters with massively-parallel architectures. In: Malyshkin, V. (ed.) PaCT 2015. LNCS, vol. 9251, pp. 436–444. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-21909-7_42
Burstedde, C., et al.: Extreme-scale AMR. In: Proceedings of the 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, pp. 1–12. IEEE Computer Society (2010)
Pavlukhin, P., Menshov, I.: GPU-aware AMR on octree-based grids. In: Malyshkin, V. (ed.) PaCT 2019. LNCS, vol. 11657, pp. 214–220. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-25636-4_17
Menshov, I., Pavlukhin, P.: GPU-native gas dynamic solver on octree-based AMR grids. J. Phys.: Conf. Ser. 1640, 012017 (2020). https://publishingsupport.iopscience.iop.org/questions/how-to-cite-an-iop-conference-series-paper/
Alhadeff, A., Leon, S.E., Celes, W., Paulino, G.H.: Massively parallel adaptive mesh refinement and coarsening for dynamic fracture simulations. Eng. Comput. 32(3), 533–552 (2016). https://doi.org/10.1007/s00366-015-0431-0
Giuliani, A., Krivodonova, L.: Adaptive mesh refinement on graphics processing units for applications in gas dynamics. J. Comput. Phys. 381, 67–90 (2019)
VanLeer, B.: Towards the ultimate conservative difference scheme V: a second-order sequel to Godunov’s method. J. Comp. Phys. 32, 101–136 (1979)
Godunov, S.K., Zabrodin, A.V., Ivanov, M.V., Kraiko, A.N., Prokopov, G.P.: Numerical solution of multidimensional problems of gas dynamics. Nauka, Moscow (1976)
Menshov, I.S., Pavlukhin, P.V.: Efficient parallel shock-capturing method for aerodynamics simulations on body-unfitted cartesian grids. Comput. Math. Math. Phys. 56(9), 1651–1664 (2016). https://doi.org/10.1134/S096554251609013X
Lohner, R.: An adaptive finite element scheme for transient problems in CFD. Comput. Methods Appl. Mech. Eng. 61, 323–338 (1987)
Dumbser, M., Zanotti, O., Hidalgo, A., Balsara, D.S.: ADER-WENO finite volume schemes with space-time adaptive mesh refinement. J. Comput. Phys. 248, 257–286 (2013)
Nguyen, H.: GPU Gems 3, 1st edn. Addison-Wesley Professional, Boston (2007)
Lax, P.D., Liu, X.-D.: Solution of two-dimensional Riemann problems of gas dynamics by positive schemes. SIAM J. Sci. Comput. 19(2), 319–340 (1998)
Sedov, L.: Similarity and Dimensional Methods in Mechanics. Academic Press, NewYork (1959)
Acknowledgments
This work was supported by Moscow Center of Fundamental and Applied Mathematics, Agreement with the Ministry of Science and Higher Education of the Russian Federation, No. 075-15-2019-1623.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Pavlukhin, P., Menshov, I. (2021). On Defragmentation Algorithms for GPU-Native Octree-Based AMR Grids. In: Malyshkin, V. (eds) Parallel Computing Technologies. PaCT 2021. Lecture Notes in Computer Science(), vol 12942. Springer, Cham. https://doi.org/10.1007/978-3-030-86359-3_18
Download citation
DOI: https://doi.org/10.1007/978-3-030-86359-3_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86358-6
Online ISBN: 978-3-030-86359-3
eBook Packages: Computer ScienceComputer Science (R0)