Optimal multidimensional accuracy-resource trade-off

PN 5-1

Project description

The goal of this project is to analyze, realize and combine unconventional numerical and high performance computing methods to develop truly adaptive simulation software with an important focus on novel optimality criteria. Together with our partner projects, the entire hardware spectrum from conventional CPU-based clusters, accelerator hardware to mobile resource-poor devices will be targeted. In this first project phase, we will optimize in the underlying multi-dimensional discrete-continuous space. This space ranges from model types and discretization over floating point accuracy and (non-)linear solver types to parallelization schemes. The core result of the project will be (a) highly efficient simulation software modules incorporating a wide range of hardware characteristics in terms of compute and communication, (b) the definition of a high-dimensional parameter space, a selection of objective functions, and initial optimization techniques. This will serve as a basis for advancing the first research question (accuracy-resource trade-off) on PN 5 into the fourth one (full accuracy-precision-resource trade-off) in a follow-up project.

Project information

Project title Optimal multidimensional accuracy-resource trade-off
Project leaders Dominik Göddeke
Miriam Schulte
Project partners Melanie Herschel (PN7-1)
Frank Leymann (PN 7-2)
Kurt Rothermel (PN7-3)
Oliver Röhrle (PN7-4)
Project duration October 2019 - March 2023
Project number PN 5-1

Publications PN 5-1

  1. 2024

    1. B. Maier, D. Göddeke, F. Huber, T. Klotz, O. Röhrle, and M. Schulte, “OpenDiHu: An Efficient and Scalable Framework for Biophysical Simulations of the Neuromuscular System,” Journal of Computational Science, vol. 79, 2024, doi: https://doi.org/10.1016/j.jocs.2024.102291.
    2. F. Huber, P.-C. Bürkner, D. Göddeke, and M. Schulte, “Knowledge-based modeling of simulation behavior for Bayesian optimization,” Computational Mechanics, vol. 74, no. 1, Art. no. 1, Jul. 2024, doi: 10.1007/s00466-023-02427-3.

  2. 2021

    1. B. Maier, D. Schneider, M. Schulte, and B. Uekermann, “Bridging Scales with Volume Coupling --- Scalable Simulations of Muscle Contraction and Electromyography,” in High Performance Computing in Science and Engineering ’21: Transactions of the High Performance Computing Center, Stuttgart (HLRS) 2021, in High Performance Computing in Science and Engineering ’21: Transactions of the High Performance Computing Center, Stuttgart (HLRS) 2021. , 2021.
    2. M. Osorno, M. Schirwon, N. Kijanski, R. Sivanesapillai, H. Steeb, and D. Göddeke, “A cross-platform, high-performance SPH toolkit for image-based flow simulations on the pore scale of porous media,” Computer Physics Communications, vol. 267, no. 108059, Art. no. 108059, Oct. 2021, doi: 10.1016/j.cpc.2021.108059.
    3. A. Krämer et al., “Multi-physics multi-scale HPC simulations of skeletal muscles,” High Performance Computing in Science and Engineering ’20: Transactions of the High Performance Computing Center, Stuttgart(HLRS) 2020, pp. 185–203, 2021, doi: 10.1007/978-3-030-80602-6_13.
    4. A. Rörich, T. A. Werthmann, D. Göddeke, and L. Grasedyck, “Bayesian inversion for electromyography using low-rank tensor formats,” Inverse Problems, vol. 37, no. 5, Art. no. 5, Mar. 2021, doi: 10.1088/1361-6420/abd85a.
To the top of the page