Advertisement

Multi-fidelity Surrogate Modeling for Application/Architecture Co-design

  • Yiming Zhang
  • Aravind NeelakantanEmail author
  • Nalini Kumar
  • Chanyoung Park
  • Raphael T. Haftka
  • Nam H. Kim
  • Herman Lam
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10724)

Abstract

The HPC community has been using abstract, representative applications and architecture models to enable faster co-design cycles. While developers often qualitatively verify the correlation of the application abstractions to the parent application, it is equally important to quantify this correlation to understand how the co-design results translate to the parent application. In this paper, we propose a multi-fidelity surrogate (MFS) approach which combines data samples of low-fidelity (LF) models (representative apps and architecture simulation) with a few samples of a high-fidelity (HF) model (parent app). The application of MFS is demonstrated using a multi-physics simulation application and its proxy-app, skeleton-app, and simulation models. Our results show that RMSE between predictions of MFS and the baseline HF models was 4%, which is significantly better than using either LF or HF data alone, demonstrating that MFS is a promising approach for predicting the parent application performance while staying within a computational budget.

Keywords

Performance estimation Multi-fidelity surrogate Behavioral emulation 

Notes

Acknowledgment

This work is supported by the U.S. Department of Energy, National Nuclear Security Administration, Advanced Simulation and Computing Program, as a Cooperative Agreement under the Predictive Science Academic Alliance Program, under Contract No. DE-NA0002378.

References

  1. 1.
    Center for Compressible Multiphase Turbulence webpage, 15 February 2015. https://www.eng.ufl.edu/ccmt/
  2. 2.
  3. 3.
    Vulcan Supercomputer, LLNL webpage. https://computation.llnl.gov/computers/vulcan
  4. 4.
    Balabanov, V., Grossman, B., Watson, L., Mason, W., Haftka, R.: Multifidelity response surface model for HSCT wing bending material weight. In: 7th AIAA/USAF/NASA/ISSMO Symposium on Multidisciplinary Analysis and Optimization, p. 4804 (1998)Google Scholar
  5. 5.
    Banerjee, T., Hackl, J., Shringarpure, M., Islam, T., Balachandar, S., Jackson, T., Ranka, S.: CMT-bone: a proxy application for compressible multiphase turbulent flows. In: 2016 IEEE 23rd International Conference on High Performance Computing (HiPC), pp. 173–182, December 2016Google Scholar
  6. 6.
    Barrett, R.F., Borkar, S., Dosanjh, S.S., Hammond, S.D., Heroux, M.A., Hu, X.S., Luitjens, J., Parker, S.G., Shalf, J., Tang, L.: On the role of co-design in high performance computing (2013)Google Scholar
  7. 7.
    Barrett, R.F., Vaughan, C.T., Heroux, M.A.: Minighost: a miniapp for exploring boundary exchange strategies using stencil computations in scientific parallel computing. Sandia National Laboratories, Technical report SAND 5294832 (2011)Google Scholar
  8. 8.
    Box, G.E.P., Hunter, J.S., Hunter, W.G.: Statistics for Experimenters: Design, Innovation, and Discovery, vol. 2. Wiley-Interscience, New York (2005)zbMATHGoogle Scholar
  9. 9.
    Dosanjh, S.S., Barrett, R.F., Doerfler, D., Hammond, S.D., Hemmert, K.S., Heroux, M.A., Lin, P.T., Pedretti, K.T., Rodrigues, A.F., Trucano, T.: Exascale design space exploration and co-design. Future Gener. Comput. Syst. 30, 46–58 (2014)CrossRefGoogle Scholar
  10. 10.
    Dosanjh, S.S., Barrett, R.F., Doerfler, D., Hammond, S.D., Hemmert, K.S., Heroux, M.A., Lin, P.T., Pedretti, K.T., Rodrigues, A.F., Trucano, T., et al.: Assessing the role of mini-applications in predicting key performance characteristics of scientific and engineering applications. J. Parallel Distrib. Comput. 30, 107–122 (2014)Google Scholar
  11. 11.
    Ellis, M., Mathews, E.: A new simplified thermal design tool for architects. Build. Environ. 36, 1009–1021 (2011)CrossRefGoogle Scholar
  12. 12.
    Fernández-Godino, M.G., Park, C., Kim, N.H., Haftka, R.T.: Review of multi-fidelity models. arXiv preprint arXiv:1609.07196 (2016)
  13. 13.
    Heroux, M.A., Doerfler, D.W., Crozier, P.S., Willenbring, J.M., Edwards, H.C., Williams, A., Rajan, M., Keiter, E.R., Thornquist, H.K., Numrich, R.W.: Improving performance via mini-applications. Sandia National Laboratories, Technical report SAND2009-5574 3 (2009)Google Scholar
  14. 14.
    Huang, D., Allen, T., Notz, W., Miller, R.: Sequential kriging optimization using multiple-fidelity evaluations. Struct. Multi. Optim. 32, 369–382 (2006)CrossRefGoogle Scholar
  15. 15.
    Jin, R., Chen, W., Sudjianto, A.: An efficient algorithm for constructing optimal design of computer experiments. J. Stat. Plan. Infer. 134, 268–287 (2005)MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Karlin, I., Keasler, J., Neely, R.: Lulesh 2.0 updates and changes. Technical report LLNL-TR-641973 (2013)Google Scholar
  17. 17.
    Kennedy, M.C., O’Hagan, A.: Bayesian calibration of computer models. J. Roy. Stat. Soc. Ser. B (Stat. Methodol.) 63(3), 425–464 (2001)MathSciNetCrossRefzbMATHGoogle Scholar
  18. 18.
    Kumar, N., Sringarpure, M., Banerjee, T., Hackl, J., Balachandar, S., Lam, H., George, A., Ranka, S.: CMT-bone: a mini-app for compressible multiphase turbulence simulation software. In: 2015 IEEE International Conference on Cluster Computing, pp. 785–792, September 2015Google Scholar
  19. 19.
    Kumar, N., Pascoe, C., Hajas, C., Lam, H., Stitt, G., George, A.: Behavioral emulation for scalable design-space exploration of algorithms and architectures. In: Taufer, M., Mohr, B., Kunkel, J.M. (eds.) ISC High Performance 2016. LNCS, vol. 9945, pp. 5–17. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-46079-6_1 CrossRefGoogle Scholar
  20. 20.
    Le Gratiet, L.: Multi-fidelity Gaussian process regression for computer experiments. Universit Paris-Diderot-Paris VII (2013)Google Scholar
  21. 21.
    Peherstorfer, B., Willcox, K., Gunzburger, M.: Survey of multifidelity methods in uncertainty propagation, inference, and optimization (2016)Google Scholar
  22. 22.
    Qian, P.Z., Wu, C.J.: Bayesian hierarchical modeling for integrating low-accuracy and high-accuracy experiments. Technometrics 50, 192–204 (2008)MathSciNetCrossRefGoogle Scholar
  23. 23.
    Xiong, Y., Chen, W., Tsui, K.L.: A new variable-fidelity optimization framework based on model fusion and objective-oriented sequential sampling. J. Mech. Des. 130(11), 111401 (2008)CrossRefGoogle Scholar
  24. 24.
    Zhang, Y., Kim, N.H., Park, C., Haftka, R.T.: Multi-fidelity surrogate based on single linear regression. ArXiv e-prints arXiv:1705.02956 (2017)
  25. 25.
    Zhang, Y., Meeker, J., Schutte, J., Kim, N., Haftka, R.: On approaches to combine experimental strength and simulation with application to open-hole-tension configuration. In: Proceedings of the American Society for Composites: Thirty-First Technical Conference (2016)Google Scholar
  26. 26.
    Zheng, L., Hendrick, T.L., Mittal, R.: A multi-fidelity modelling approach for evaluation and optimization of wing stroke aerodynamics in flapping flight. J. Fluid Mech. 721, 118–154 (2013)CrossRefzbMATHGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Yiming Zhang
    • 1
  • Aravind Neelakantan
    • 2
    Email author
  • Nalini Kumar
    • 2
  • Chanyoung Park
    • 1
  • Raphael T. Haftka
    • 1
  • Nam H. Kim
    • 1
  • Herman Lam
    • 2
  1. 1.Department of Mechanical and Aerospace EngineeringUniversity of FloridaGainesvilleUSA
  2. 2.Department of Electrical and Computer EngineeringUniversity of FloridaGainesvilleUSA

Personalised recommendations