A Caltech Library Service

Efficient data structures for model-free data-driven computational mechanics

Eggersmann, Robert and Stainier, Laurent and Ortiz, Michael and Reese, Stefanie (2021) Efficient data structures for model-free data-driven computational mechanics. Computer Methods in Applied Mechanics and Engineering, 382 . Art. No. 113855. ISSN 0045-7825. doi:10.1016/j.cma.2021.113855.

[img] PDF - Submitted Version
See Usage Policy.


Use this Persistent URL to link to this item:


The data-driven computing paradigm initially introduced by Kirchdoerfer & Ortiz (2016) enables finite element computations in solid mechanics to be performed directly from material data sets, without an explicit material model. From a computational effort point of view, the most challenging task is the projection of admissible states at material points onto their closest states in the material data set. In this study, we compare and develop several possible data structures for solving the nearest-neighbor problem. We show that approximate nearest-neighbor (ANN) algorithms can accelerate material data searches by several orders of magnitude relative to exact searching algorithms. The approximations are suggested by—and adapted to—the structure of the data-driven iterative solver and result in no significant loss of solution accuracy. We assess the performance of the ANN algorithm with respect to material data set size with the aid of a 3D elasticity test case. We show that computations on a single processor with up to one billion material data points are feasible within a few seconds execution time with a speed up of more than 10⁶ with respect to exact k-d trees.

Item Type:Article
Related URLs:
URLURL TypeDescription Paper
Stainier, Laurent0000-0001-6719-6616
Ortiz, Michael0000-0001-5877-4824
Additional Information:© 2021 Published by Elsevier B.V. Received 30 November 2020, Revised 27 March 2021, Accepted 7 April 2021, Available online 6 May 2021. MO gratefully acknowledges the support of the Deutsche Forschungsgemeinschaft (DFG), Germany through the Sonderforschungsbereich 1060 “The mathematics of emergent effects” and the Mercator fellowhip within the collaborative research centre SFB/TRR87. SR and RE gratefully acknowledge the financial support of the Deutsche Forschungsgemeinschaft (DFG), Germany through the project RE 1057/40-2 “Model order reduction in space and parameter dimension — towards damage-based modeling of polymorphic uncertainty in the context of robustness and reliability” within the priority program SPP 1886 “Polymorphic uncertainty modelling for the numerical design of structures”. Finally, all authors acknowledge the financial support of the DFG, Germany and French Agence Nationale de la Recherche (ANR) through the project “Direct Data-Driven Computational Mechanics for Anelastic Material Behaviours” (project numbers: ANR-19-CE46-0012-01, RE 1057/47-1) within the French-German Collaboration for Joint Projects in Natural, Life and Engineering (NLE) Sciences. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Funding AgencyGrant Number
Deutsche Forschungsgemeinschaft (DFG)SFB 1060
Deutsche Forschungsgemeinschaft (DFG)RE 1057/40-2
Deutsche Forschungsgemeinschaft (DFG)SPP 1886
Agence Nationale pour la Recherche (ANR)ANR-19-CE46-0012-01
Deutsche Forschungsgemeinschaft (DFG)RE 1057/47-1
Subject Keywords:Data-driven computing; Solid mechanics; Nearest neighbor problem; Approximate nearest-neighbor search; Data structures; Data science
Record Number:CaltechAUTHORS:20210225-132741960
Persistent URL:
Official Citation:Robert Eggersmann, Laurent Stainier, Michael Ortiz, Stefanie Reese, Efficient data structures for model-free data-driven computational mechanics, Computer Methods in Applied Mechanics and Engineering, Volume 382, 2021, 113855, ISSN 0045-7825, (
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:108211
Deposited By: George Porter
Deposited On:01 Mar 2021 15:17
Last Modified:06 May 2021 21:13

Repository Staff Only: item control page