Skip to main content
SpringerLink
Log in

JXPAMG: a parallel algebraic multigrid solver for extreme-scale numerical simulations

  • Regular Paper
  • Published:
CCF Transactions on High Performance Computing Aims and scope Submit manuscript

Abstract

JXPAMG is a parallel algebraic multigrid (AMG) solver for solving the extreme-scale, sparse linear systems on modern supercomputers. JXPAMG features the following characteristics: 1) It integrates some application-driven parallel AMG algorithms, including αSetup-AMG (adaptive Setup based AMG), AI-AMG (algebraic interface based AMG) and AMG-PCTL (physical-variable based coarsening two-level AMG); 2) A hierarchical parallel sparse matrix data structure, labeled hierarchical parallel Compressed Sparse Row (hpCSR), that matches the computer architecture is designed, and the highly scalable components based on hpCSR are implemented; 3) A flexible software architecture is designed to separate algorithm development from implementation. These characteristics allow JXPAMG to use different AMG strategies for different application features and architecture features, and thereby JXPAMG becomes aware of changes in these features. This paper introduces the algorithms, implementation techniques and applications of JXPAMG. Numerical experiments for typical real applications are given to illustrate the strong and weak parallel scaling properties of JXPAMG.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

References

  • Adams, M., Bayraktar, H., Keaveny, T., Papadopoulos, P.: Applications of algebraic multigrid to large-scale finite element analysis of whole bone micro-mechanics on the IBM SP, SC'03, 2003.

  • Adams, M., Bayraktar, H., Keaveny, T., Papadopoulos, P.: Ultra scalable implicit finite element analysis in solid mechanics with over a half billion DOFs, SC'04, 2004.

  • Baker, A., Falgout, R., Kolev, T., Yang, U.: Multigrid smoothers for Ultra-Parallel computing. SIAM J. Sci. Comput. 33, 2864–2887 (2011)

    Article  MathSciNet  MATH  Google Scholar 

  • Baldwin, C., Brown, P., Falgout, R., Graziani, F., Jones, J.: Iterative linear solvers in 2D radiation-hydrodynamics code: methods and performance. J. Comput. Phys. 154, 1–40 (1999)

    Article  MATH  Google Scholar 

  • Brandt, A., McCormick, S., Ruge, J.: Algebraic Multigrid (AMG) for automatic solutioin with application to geodetic computations, Institute for Computational Studies, POB 1852. Fort Collins, Colorado (1982)

    Google Scholar 

  • Brezina, M., Tong, C., Becker, R.: Parallel algebraic multigrids for structural mechanics. SIAM J. Sci. Comput. 27(5), 1534–1554 (2006)

    Article  MathSciNet  MATH  Google Scholar 

  • Cleary, A., Falgout, R., Henson, V., Jones, J.: Coarse-grid selection for parallel algebraic multigrid. Lect. Notes Comput. Sci. 1457, 104–115 (1998)

    Article  MathSciNet  Google Scholar 

  • Clees, T., Stüben, K.: Algebraic multigrid for industrial semiconductor device simulation, Proceedings of the 1st International Conference on Challenges in Scientific Computing, Berlin, Germany (2002)

  • Clees, T., Ganzer, L.: An efficient algebraic multigrid solver strategy for adaptive implicit methods in oil-reservoir simulation. SPE J. 15(3), 670–681 (2010)

    Article  Google Scholar 

  • CORAL-2 Benchmarks: https://asc.llnl.gov/coral-2-benchmarks.

  • de Sterck, H., Yang, U., Heys, J.: Reducing complexity in parallel algebraic multigrid preconditioners. SIAM J. Matrix Anal. Appl. 27(4), 1019–1039 (2006)

    Article  MathSciNet  MATH  Google Scholar 

  • Falgout, R., Schroder, J.: Non-Galerkin coarse grids for algebraic multigrid. SIAM J. Sci. Comput. 36, C309–C334 (2014)

    Article  MathSciNet  MATH  Google Scholar 

  • Falgout, R., Brannick, J., Brezina, M., Manteuffel, T., McCormick, S.: New multigrid solver advances in TOPS. J. Phys.: Conf Ser. 16, 456–460 (2005)

    Google Scholar 

  • Füllenbach, T., Stüben, K., Mijalkovic, S.: Application of an algebraic multigrid solver to process simulation problems, Proceedings of the IEEE Intern. Conference on Simulation of Semiconductor Processes and Devices, Seattle(WA), USA, 225–228 (2000)

  • Gries, S., Stüben, K., Brown, G., Chen, D., Collins, D.: Preconditioning for efficiently applying algebraic multigrid in fully implicit reservoir simulations. SPE J. 19(4), 726–736 (2014)

    Article  Google Scholar 

  • Henson, V.: New directions for algebraic multigrid: solutions for large scale multiphysics problems, LLNL Report No.UCRL-ID-151775 (2003)

  • Henson, V., Yang, U.: BoomerAMG: a parallel algebraic multigrid solver and preconditioner. Appl. Numer. Math. 41, 155–177 (2002)

    Article  MathSciNet  MATH  Google Scholar 

  • Heroux, M., Carter, J., Thakur, R., Vetter, J., McInners, L., Ahrens, J., Robert, J.: ECP Software Technology Capability Assessment Report-Public V1.5, (2019)

  • HYPRE Web page: https://computing.llnl.gov/projects/hypre-scalable-linear-solvers-multigrid-methods.

  • Joubert, W., Cullum, J.: Scalable algebraic multigrid on 3500 processors, Electron. Trans. Numer. Anal., 23: 105–128, 2006. Also appear at Technical Report No.LAUR03–568, LANL, 2003.

  • Liu, Q., Mo, Z., Zhang, A., Yang, Z.: JAUMIN: a programming framework for large-scale numerical simulation on unstructured meshes. CCF Trans. HPC 1, 35–48 (2019)

    Article  Google Scholar 

  • McCormick, S., Ruge, J.: Algebraic multigrid methods applied to problems in computational structural mechanics, Technical Report 19900060129, NASA. Also available in State-of-the-art surveys on computational mechanics, 237–270, New York, American Society of Mechnical Engineers, 1989.

  • Mo, Z., Xu, X.: Relaxed RS0 or CLJP coarsening strategy for parallel AMG methods. Parallel Comput. 33(3), 174–185 (2007)

    Article  MathSciNet  Google Scholar 

  • Mo, Z., Zhang, A., Cao, X., Liu, Q., Xu, X., An, H., Pei, W., Zhu, S.: JASMIN: a parallel software infrastructure for scientific computing. Front. Comput. Sci. China 4(4), 480–488 (2010)

    Article  Google Scholar 

  • MueLu Web page: https://trilinos.github.io/muelu.html.

  • Pei, W.: The construction of simulation algorithm for laser fusion. Commun. Comput. Phys. 2(2), 255–270 (2007)

    Google Scholar 

  • PETSc Web page: https://petsc.org/.

  • Ruge, J.: AMG for problems of elasticity. Appl. Math. Comput. 19, 293–309 (1986)

    MATH  Google Scholar 

  • Ruge, J., Stüben, K.: Algebraic Multigrid, in Multigrid Methods, Frontiers Appl. Math. 3, S. McCormick (ed.), SIAM, Philadelphia, 73–130, 1987.

  • Saad, Y.: Iterative methods for sparse linear systems, 2nd edn. SIAM, Philadelphia (2003)

    Book  MATH  Google Scholar 

  • SAMG Web page: https://www.scai.fraunhofer.de/en/business-research-areas/fast-solvers/products/samg.html.

  • Sellappan, V., Desam, B.: Accelerating ANSYS fluent simulations with NVIDIA GPUs. ANSYS Advantage 1X(1), 51–53 (2015)

    Google Scholar 

  • Stüben, K.: Algebraic multigrid (AMG): an introduction with applications, In: Multigrid, U. Trottenberg, C. Oosterlee, A. Schuller, Academic Press, 413–532, (2001)

  • Stüben, K., Delaney, P., Chmakov, S.: Algebraic multigrid for ground water flow and oil reservoir simulation, Proceedings of the Conference “MODFLOW and More 2003: Understanding through Modeling”, Intern. Ground Water Modeling Center (IGWMC) (2003)

  • Stüben, K., Ruge, J., Clees, T., Gries, S., Plum, J.: Algebraic Multigrid - From Academia to Industry, in M. Griebel, A. Schuller, M. Schweitzer (eds.), Scientific Computing and Algorithms in Industrial Simulations, Springer, 2017.

  • Tian, R., Zhou, M., Wang, J., Li, Y., An, H., Xu, X., Wen, L., Wang, L., Xu, Q., Leng, J., Xu, R., Zhang, B., Liu, W., Mo, Z.: A challenging dam structural analysis: large-scale implicit thermo-mechanical coupled contact simulation on Tianhe-II. Comput. Mech. 63, 99–119 (2019)

    Article  MathSciNet  Google Scholar 

  • Trilinos Web page: https://trilinos.github.io/.

  • Vanek, P., Mandel, J., Brezina, M.: Algebraic multigrid by smoothed aggregation for second and fourth order elliptic problems. Computing 56, 179–196 (1996)

    Article  MathSciNet  MATH  Google Scholar 

  • Vanek, P., Brezina, M., Mandel, J.: Convergence of algebraic multigrid based on smoothed aggregation. Numer. Math. 88, 559–579 (2001)

    Article  MathSciNet  MATH  Google Scholar 

  • Xiao, Y., Shu, S., Zhang, P., Mo, Z., Xu, J.: Semi-coarsening algebraic multigrid methods for 2D–3T energy equations. J. Num. Method Comp. Appl. 24, 293–303 (2003)

    Google Scholar 

  • Xu, X.: Parallel algebraic multigrid methods: state-of-the art and challenges for extreme-scale applications. J. Num. Method Comp. Appl. 40(4), 243–260 (2019)

    MathSciNet  MATH  Google Scholar 

  • Xu, X., Mo, Z.: Algebraic interface based coarsening AMG preconditioner for multi-scale sparse matrices with applications to radiation hydrodynamics computation. Numer. Linear Algebra Appl. 24(2), e2078 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  • Xu, X., Mo, Z., An, H.: An algebraic two-level method for 2D–3T radiation diffusion equations. Chinese J. Comput. Phys. 26(1), 1–8 (2009)

    Google Scholar 

  • Xu, X., Mo, Z., An, H.: An adaptive AMG preconditioning strategy for solving large-scale sparse linear systems. Sci. Sin. Inf. 46(10), 1411–1420 (2016)

    Article  Google Scholar 

  • Xu, X., Mo, Z., Yue, X., An, H., Shu, S.: αSetup-AMG: an adaptive-setup-based parallel AMG solver for sequence of sparse linear systems. CCF Trans. HPC 2, 98–110 (2020)

    Article  Google Scholar 

  • Yang, U.: On the use of relaxation parameters in hybrid smoothers. Numer. Linear Algebra Appl. 11, 155–172 (2004)

    Article  MathSciNet  MATH  Google Scholar 

  • Yue, X., Shu, S., Xu, X., Zhou, Z.: An adaptive combined preconditioner with applications in radiation diffusion equations. Commun. Comput. Phys. 18(5), 1313–1335 (2015)

    Article  MathSciNet  MATH  Google Scholar 

  • Yue, X., Xu, X., Shu, S.: JASMIN-based two-dimensional adaptive combined preconditioner for radiation diffusion equations in inertial fusion research. East Asian J. Appl. Math. 7, 495–507 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  • Yue, X., Zhang, S., Xu, X., Shu, S., Shi, W.: Algebraic multigrid block preconditioning for multi-group radiation diffusion equations. Commun. Comput. Phys. 29(3), 831–852 (2021)

    Article  MathSciNet  MATH  Google Scholar 

  • Zhou, Z., Xu, X., Shu, S., Feng, C., Mo, Z.: Adaptive two-level preconditioners for 2D–3T radiation diffusion equations. Chinese J. Comput. Phys. 29(4), 475–483 (2012)

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Laboratory of Computational Physics, Institute of Applied Physics and Computational Mathematics, Beijing, 100094, China

    Xiaowen Xu, Silu Huang, Shaoliang Hu & Zeyao Mo

  2. National Center for Applied Mathematics in Hunan, Key Laboratory of Intelligent Computing & Information Processing of Ministry of Education, Hunan Key Laboratory for Computation and Simulation in Science and Engineering, Xiangtan University, Xiangtan, 411105, China

    Xiaoqiang Yue, Chunsheng Feng & Shi Shu

  3. Graduate School of China Academy of Engineering Physics, Beijing, China

    Runzhang Mao, Yuntong Deng, Haifeng Zou & Xiao Liu

  4. CAEP Software Center for High Performance Numerical Simulation, Beijing, China

    Xiaowen Xu, Shaoliang Hu & Zeyao Mo

Authors
  1. Xiaowen Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Xiaoqiang Yue
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Runzhang Mao
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Yuntong Deng
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Silu Huang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Haifeng Zou
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Xiao Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Shaoliang Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Chunsheng Feng
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Shi Shu
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Zeyao Mo
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Xiaowen Xu.

Ethics declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Additional information

This work is supported by the Science Challenge Project (TZZT2019) and NSFC (62032023, 11971414).

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xu, X., Yue, X., Mao, R. et al. JXPAMG: a parallel algebraic multigrid solver for extreme-scale numerical simulations. CCF Trans. HPC 5, 72–83 (2023). https://doi.org/10.1007/s42514-022-00125-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s42514-022-00125-9

Keywords

Search

Navigation